Towards a Euro-Rwanda policy? The proposed new EU asylum law rules on ‘safe third countries’

 

Professor Steve Peers, Royal Holloway University of London

Photo credit: Home Office, via Wikimedia Commons – the former Home Secretary meets Rwanda’s foreign minister

Introduction

An unlawful attempt to remove asylum-seekers to an unsafe country that they had no connection with, in breach of basic human rights – as confirmed by the Supreme Court. A desperate attempt to overturn the Supreme Court ruling by an Act of Parliament (the ‘Safety’ of Rwanda Act), declaring the dodgy country to be ‘safe’, regardless of reality. Ultimately, the cancellation of the policy by a new government, the previous government having squandered hundreds of millions of pounds on its unlawful obsession.

The flaws with the UK’s Rwanda policy were so huge, they must have been visible from outer space. But were they visible to the European Union?

Yesterday’s proposal from the EU Commission to amend EU asylum law would provide, in effect, for the development of a form of Euro-Rwanda policy, by widening the concept of ‘safe third country’ to include countries that asylum-seekers have no link with whatsoever. We don’t know yet whether EU Member States and the European Parliament will agree to it, but even tabling the proposal raises fundamental questions. Has the EU learned the lessons of the UK’s failed policy? Or is it poised to follow the previous UK government, lemming-like, by leaping into the same financial, legal and moral abyss?

EU legal framework

Current legislation

The definition of ‘safe third country’ for the EU is currently set out in Article 38 of the EU’s asylum procedures Directive. (This is about the ‘safety’ of non-EU countries, ie it is distinct from the EU Member States regarding each other as ‘safe’ countries, under the Dublin rules). It’s only an option for Member States to apply these rules, not a requirement; the Commission’s staff working document alongside the new proposal details which Member States use the concept, and how they use it.

First, Article 38(1) defines what makes a country ‘safe’. The ‘safe third country’ concept can ‘only’ be applied if Member States are satisfied that certain ‘principles’ are ensured for asylum-seekers in that country: (a) ‘life and liberty are not threatened on account of race, religion, nationality, membership of a particular social group or political opinion’; (b) ‘there is no risk of serious harm as defined in’ the EU qualification Directive; (c) ‘the principle of non-refoulement in accordance with the Geneva [Refugee] Convention is respected’; (d)  ‘prohibition of removal, in violation of the right to freedom from torture and cruel, inhuman or degrading treatment as laid down in international law, is respected’; and (e) ‘the possibility exists to request refugee status and, if found to be a refugee, to receive protection in accordance with the Geneva Convention’. According to the CJEU, Hungarian law breached the Directive, as it did not set out these guarantees fully (Cases C‑564/18, paras 29-51 of the judgment; Joined Cases C-924/19 and C-925/19 PPU, paras 148-165; and Case C‑821/19).

Secondly, Article 38(2) requires Member States to subject the application of the ‘safe third country’ rule to national law rules, including: (a) ‘requiring a connection between the applicant and the third country concerned on the basis of which it would be reasonable for that person to go to that country’; (b) ‘rules on the methodology by which the competent authorities satisfy themselves that the safe third country concept may be applied to a particular country or to a particular applicant. Such methodology shall include case-by-case consideration of the safety of the country for a particular applicant and/or national designation of countries considered to be generally safe’; and (c) ‘rules in accordance with international law, allowing an individual examination of whether the third country concerned is safe for a particular applicant which, as a minimum, shall permit the applicant to challenge the application of the safe third country concept on the grounds that the third country is not safe in his or her particular circumstances’. Also, an asylum-seeker must ‘be allowed to challenge the existence of a connection between him or her and the third country’, as referred to in point (a). According to the CJEU in the cases mentioned above, mere transit through a country was not sufficient to create a ‘connection’.

Next, Article 38(4) addresses what happens if the supposedly ‘safe’ country does not readmit the asylum-seeker:

Where the third country does not permit the applicant to enter its territory, Member States shall ensure that access to [the asylum] procedure is given in accordance with the basic principles and guarantees described in Chapter II. [ie an ‘ordinary’ examination of the merits of an asylum claim]

Interpreting this guarantee, last year the CJEU ruled that asylum applications could not be found inadmissible where the supposedly ‘safe’ country was refusing to readmit the asylum-seekers.

There are also special rules for unaccompanied minors: Article 25(6) of the Directive permits Member States to apply the current version of the special border procedure to unaccompanied minors in ‘safe third countries’ cases (among others); but in any case, they may only apply the ‘safe third country’ rule to unaccompanied minors if that is ‘in the minor’s best interests’.

The practical relevance of applying the ‘safe third country’ notion in the current law is (among other things) that it is a ground of inadmissibility under Article 33 of the Directive. And if the case is inadmissible, the asylum application is not further considered on the merits and has failed in that Member State (subject to an appeal of the inadmissibility decision, which has suspensive effect in ‘safe third countries’ cases). If any appeal fails, the failed asylum-seeker can then deprived of support as an asylum seeker under the reception conditions Directive and detained and removed under the Returns Directive. Of course, the impact of this is in theory ameliorated in ‘safe third country’ cases because the supposedly ‘safe’ country will readmit the person concerned and consider their asylum application properly – although this does not always happen in practice. ‘Safe third country’ cases can also be dealt with in the current version of the special border procedure, subject to further conditions.

Future rules: the asylum pact

The ‘safe third country’ rule in EU asylum law is already set to be amended once the asylum procedures Regulation, which is a major part of the asylum pact, becomes applicable, to applications made from June 2026. Yesterday’s proposal would amend that asylum pact law, but we can only understand the proposal’s impact in light of what the asylum pact law provides for.

First of all, although the 2024 Regulation overall entails more harmonisation of national law on asylum procedure, it leaves the ‘safe third country’ rule optional for Member States.

Secondly, Article 59(1) of that Regulation retains essentially the same definition as in the current Directive about the ‘safety’ of the country concerned, except that there is no longer a requirement that the country ensure the possibility of obtaining status under the Refugee Convention; it is sufficient that the asylum-seeker could seek ‘effective protection’ as further defined in the Regulation (Article 57): this means that either the Refugee Convention applies, or at least the person concerned has the chance to remain on the territory with subsistence support, health-care, education and ‘effective protection’ until a ‘durable solution’ is found.

The preamble clarifies the meaning of ‘effective protection’:

access to means of subsistence sufficient to maintain an adequate standard of living should be understood as including access to food, clothing, housing or shelter and the right to engage in gainful employment, for example through access to the labour market, under conditions not less favourable than those for non-nationals of the third country generally in the same circumstances.

Next, in place of a reference to national law as regards the methodology of defining ‘safe third countries’, there is a standard rule (Article 59(3)):

The assessment of whether a third country may be designated as a safe third country in accordance with this Regulation shall be based on a range of relevant and available sources of information, including information from Member States, the Asylum Agency, the European External Action Service, the United Nations High Commissioner for Refugees, the Council of Europe and other relevant international organisations.

The application of the ‘safe third country’ rule is still limited by the requirement of an individual examination and showing a connection with the country, ie it ‘may only be applied provided that’ (Article 59(5)):

a)       the applicant cannot provide elements justifying why the concept of safe third country is not applicable to him or her, in the framework of an individual assessment;

b)      there is a connection between the applicant and the third country in question on the basis of which it would be reasonable for him or her to go to that country.

As regards a ‘connection’, the preamble states that:

The connection between the applicant and the safe third country could be considered established in particular where members of the applicant’s family are present in that country or where the applicant has settled or stayed in that country.

Furthermore, the preamble goes on to state that the ‘safe third country’ principle should not be applied to anyone with rights under EU free movement law, or the EU’s family reunion Directive.

The applicant must still be given access to the procedure if the supposedly ‘safe’ country does not admit or readmit them (Article 59(9); Article 38(1)(b) likewise says that the application cannot be inadmissible on ‘safe third country’ grounds, if ‘it is clear that’ they will not be admitted or readmitted to that country’). The current information and documentation requirements are maintained (Article 59(8)).

As for unaccompanied minors, they are still not exempted from ‘safe third country’ rules, but there are additional conditions before the rule can apply to them (Article 58(6)):

A third country may only be considered to be a safe third country for an unaccompanied minor where it is not contrary to his or her best interests and where the authorities of Member States have first received from the authorities of the third country in question the assurance that the unaccompanied minor will be taken in charge by those authorities and that he or she will immediately have access to effective protection as defined in Article 57.

The preamble states that when considering applying the rule to unaccompanied minors, authorities should ‘in particular’ examine ‘the availability of sustainable appropriate care and custodial arrangements’. Interpreting a similar provision in the Returns Directive, the CJEU has ruled that unaccompanied minors cannot be removed unless these guarantees are provided for. Moreover, it should be noted that under the asylum pact it will not be possible to subject unaccompanied minors to the new version of the borders procedure on ‘safe third country’ grounds.

In addition to retaining or amending the existing ‘safe third country’ rules, there are several new relevant elements in the asylum procedures Regulation. First of all, Article 59(2) now provides that a non-EU country can be designated as ‘safe…with exceptions for specific parts of its territory or clearly identifiable categories of persons’. On this point, the CJEU has recently confirmed that, as regards the principle of ‘safe countries of origin’, the current law does not allow exceptions for parts of a country; conversely an Advocate-General’s opinion in a pending case (Alace) argues that the current law does allow for that principle to apply with exceptions for groups of people. We do not yet have a judgment in the latter case; but presumably the interpretation in the former judgment (and the latter judgment, when we have it) applies by analogy to the ‘safe third country’ rule. So to allow Member States to use these exceptions already as regards both ‘safe country’ rules, the Commission has separately already proposed the early application of these provisions of the Regulation. It remains to be seen if this proposal is agreed. The asylum procedures Regulation also allows the ‘safe third country’ rule to be applied to an individual applicant (Article 59(4)(b)); the Commission has not proposed to apply this bizarre rule early.

Next, the Regulation now provides that the EU and a particular non-EU country can agree a treaty which provides that ‘migrants admitted under that agreement will be protected in accordance with the relevant international standards and in full respect of the principle of non-refoulement’. In that case the existence of a ‘safe third country’ will be presumed – but this is ‘without prejudice’ to the guarantees for unaccompanied minors, a ‘connection’ requirement and an individual examination.

It will also be possible to have a common EU list of ‘safe third countries’, adopted by means of a further legislative amendment (Article 60); but removals of countries from that common list due to changes in circumstances can be fast-tracked (Article 63), and such removals from the list will limit Member States from designating that country as ‘safe’ for two years (Article 64).

As for the broader impact of the ‘safe third country’ rule in the asylum pact, such applications remain inadmissible; but the 2024 Regulation now specifies a time limit to make decisions on inadmissible cases (two months). Appeals in ‘safe third country’ cases still have suspensive effect. The revised version of the special border procedure (including a new deadline to decide on applications within 12 weeks, including appeals) will apply to ‘safe third country’ cases (except for unaccompanied minors, as noted already); and the asylum pact explicitly provides that being subject to the border procedure is a ground for detention.

The new proposal

When negotiating the asylum procedures Regulation, the more liberal MEPs fought off attempts to make the ‘safe third country’ rules broader – but there was a catch. The trade-off was a clause providing for a review of these rules by June 2025, by which point the European Parliament had moved to the right after its 2024 election. Hence yesterday’s (slightly early) proposal.

The new proposal will not apply to Denmark (so Denmark’s own Rwanda policy is irrelevant). Ireland can choose whether to opt in or out; for Ireland, the proposal is particularly relevant to its designation of the UK as a ‘safe third country’.

As for the content, first of all, the proposal leaves a number of provisions of the asylum pact ‘safe third country’ rules unchanged. The principle is still optional for Member States; there are no changes to the definition of ‘safety’; the procedure to designate countries as ‘safe’ is the same; there must still be an individual examination; the rule still cannot be applied if the country concerned will not admit or readmit the asylum-seeker; the conditions for applying the rule to unaccompanied minors still apply; there are still possible exceptions for parts of a country or groups of people; the rule can still be targeted on individuals; there is still a possibility for treaties between the EU and non-EU countries and for a common EU list (the Commission has not proposed to use either clause); and the rules on inadmissibility and border procedures still apply.  

However, there are two changes to the rules. The first change is to add to the ‘connection’ criterion for designation. A country could also be designated as ‘safe’ if:

-          the applicant has transited through the third country concerned; or

-          there is an agreement or an arrangement with the third country concerned requiring the examination of the merits of the requests for effective protection made by applicants subject to that agreement or arrangement

Further provisions on unaccompanied minors would specify that when applying any of these conditions (including the ‘connection’ clause) ‘the best interests of the child shall be a primary consideration’. This adds nothing to the ‘best interests’ clause already in Article 58(6). But more concretely, the proposal exempts unaccompanied minors from the possibility of being sent to a country that they neither have a connection with nor transited through. Although this does leave the new prospect of sending an unaccompanied minor to a country that they have only transited through, this new possibility will be subject to the safeguard in the 2024 Regulation of an ‘assurance that the unaccompanied minor will be taken in charge by those authorities and that he or she will immediately have access to effective protection’.

The proposal requires Member States to inform the Commission and other Member States about arrangements or agreements they enter into; it is not clear whether such an agreement or arrangement could be negotiated with the EU as a whole, or how it might relate to countries which the EU might sign with non-EU countries, referred to in Article 58(7).

The second change is to drop the requirement of suspensive effect of appeals in ‘safe third country’ cases. This would still leave asylum-seekers in such cases with a fallback guarantee: the Regulation states that for appeals without suspensive effect, there must still be at least five days for an asylum-seeker to request a court to grant suspensive effect. The asylum-seeker cannot be removed during that grace period, or pending the court’s decision on such a request.

Although, as noted already, the proposal would not change the rules on inadmissibility or border procedures, it would mean that a greater percentage of asylum-seekers might fall within the scope of such rules, given the additional scope to apply the ‘safe third country’ principle.

Comments

In addition to adding a transit criterion to the ‘safe third country’ rule, which would particularly important to the EU’s near neighbours (including the UK, given that asylum-seekers sometimes transit the UK on the way to Ireland), the proposal would add a fully-fledged ‘Rwanda clause’ to EU asylum law: asylum-seekers could be sent to a country which they neither have a connection with nor have transited to.

The text of the proposal encompasses both versions of the previous UK government’s Rwanda policy: ‘agreement or an arrangement’ covers both the treaty which the UK and Rwanda agreed after the Supreme Court struck the policy down, and the more informal arrangements agreed before that.

On the other hand, unlike the UK’s Rwanda policy, there is no rule in EU law that the applications from asylum-seekers who entered illegally are inadmissible; that is not a ground for inadmissibility of an asylum application under EU law (even after this proposal), and indeed EU law still requires that the ‘safe third country’ rule (in any EU version) cannot apply unless the country concerned admits or readmits the asylum-seeker. Equally, as confirmed by the CJEU, an application cannot be regarded as inadmissible in such circumstances. Logically, this should apply where (as in the UK/Rwanda situation) a supposedly ‘safe’ country obviously lacks the capacity to take all the asylum-seekers that might theoretically be sent to it under a treaty – even if the numbers that could be sent there are nominally uncapped by the treaty – especially where the treaty leaves that country with a discretionary power to accept or reject any asylum-seeker that the other country might want to send there.

Nor does the EU’s Regulation – or any national law implementing it – benefit from the UK principle of parliamentary sovereignty, which would have presumably protected the UK’s Safety of Rwanda Act – which deemed Rwanda to be safe despite the UK Supreme Court ruling – from being struck down by the courts. While the Regulation does not expressly state that the designation of a ‘safe third country’ can be challenged as such, it would surely be a breach of Article 47 of the EU Charter of Fundamental Rights to prevent such a challenge. (By analogy, the most recent CJEU judgment on ‘safe countries of origin’ (discussed above) said that Article 47 of the Charter applied to such challenges; and the pending Alace case has implicitly raised this question again). Certainly, it is hard to imagine the CJEU, or (one would hope) national courts in the EU, accepting anything like the UK’s explicit ban on any judicial review of the designation of ‘safety’, coupled with (mostly) disapplying national and international human rights law.

In contrast, both the Regulation and the UK’s Act do allow asylum-seekers to challenge whether a country is ‘safe’ in their particular circumstances – although the EU’s version of this possibility (‘elements justifying why the concept of safe third country is not applicable to him or her’) appears rather broader than the UK’s (‘compelling evidence relating specifically to the person’s particular individual circumstances’).

The EU proposal is broader than the failed UK policy in one respect: it is not limited to illegal entrants. The only asylum-seekers excluded from the possible application of a Euro-Rwanda treaty are unaccompanied minors. (Member States might be tempted to ‘time out’ their protection, ie send them to Rwanda etc as soon as they turn 18; but there are deadlines to decide on asylum applications in EU law, and CJEU case law has rejected attempts to ‘time out’ unaccompanied minors in other asylum contexts, focussing on how old they were when they applied for asylum)

Another obvious, and fundamental, issue is the willingness of non-EU countries to accept those who have transited through the territory, or to enter into agreements or arrangements as referred to in the proposal. The previous UK government travelled the world – falsely briefing the press along the way – before it found a foreign government willing to do an asylum deal with it. Even then, and despite sending huge sums of British taxpayer money to Rwanda, there were questions about the capacity there compared to the numbers of asylum-seekers who could have been subject to the UK’s Rwanda policy. Will EU Member States have any greater success finding a non-EU country that is willing and able to take their asylum-seekers – and which is also genuinely safe? (One would expect any Member State literally doing a deal with Rwanda itself to be challenged in court, with the challengers referring to the arguments which persuaded the UK Supreme Court)

It should be recalled that countries have no underlying obligation to take back non-citizens who have transited their territory – and still less to take those who have never been anywhere near it. They might agree to a treaty which requires them to admit people who passed through the territory, or whom they gave a visa or residence permit too, as EU readmission treaties with non-EU countries usually require (see the EU-Albania readmission treaty, for instance). But even then there is a question of evidence to prove that they were there; and the non-EU countries concerned only agreed to these treaties on the basis of a quid pro quo (such as visa facilitation, or the promise of eventual visa waivers). While the EU’s readmission treaties would be relevant to the first new ground of defining ‘safe third countries’ (ie transit countries), they would obviously not be relevant to the second new ground, as they do not include rules on processing asylum applications.

(By the way, ‘Rwanda clauses’ obviously cannot be defended based on the often-heard claim that ‘asylum seekers ought to apply in the first safe country’. By definition, Rwanda policies apply to countries that the asylum-seeker has not passed through, and probably was never even close to: the journey from (say) Afghanistan to the EU or the UK passes nowhere near Rwanda.)

Compared to the ‘safe third country’ proposals for non-EU countries, the EU’s Dublin system is built upon mutual trust between EU Member States, which are subject to many human rights obligations (and, for most Member States, harmonised rules of EU asylum law). It has criteria to determine which Member State is responsible for an asylum application; binds Member States to accept asylum-seekers they are responsible for under the rules; and includes detailed provisions on the process of transferring asylum-seekers to the responsible Member State, including rules on evidence. And even then, there are significant problems with applying the Dublin system in practice – including as regards the principle of mutual trust.

None of these features are present between EU Member States and non-EU countries, and it would take some time to develop them – on top of the issue of political willingness of the non-EU countries to sign up, and the fundamental question of whether those countries are ‘safe’ at all. While the Commission’s proposal does not reproduce all the features of the UK’s failed policy, it is similar enough to raise comparable questions about its feasibility, and – if Member States also try to override court rulings about the safety of the countries concerned – its legality and morality too.

Friends with benefits: the legal elements of the reset of the EU/UK relationship

 

Professor Steve Peers, Royal Holloway University of London

Photo credit: Julian Herzog, via Wikimedia Commons

Introduction

Advocates of Quebec’s independence from Canada have long argued that their plan would not entail complete separation from Canada, but rather a fusion of independence and continuing links that they described as ‘sovereignty-association’. Conversely, opponents of Quebec separation derided this suggestion as ‘divorce with bed privileges’. Why should one party to a marriage get to walk away – but still expect sex?

On the other hand…what if sex was never the problem? What if both sides to the relationship could agree to end the marriage as a whole, yet consent to occasional mutually enjoyable booty calls? Especially if the relationship with that cute American that one of them was always fantasising about while the couple was married was not really working out as planned.

The debate about the post-Brexit UK/EU relationship has raised similar issues, albeit with less horny analogies: ‘cherry-picking’ and ‘having one’s cake and eating it’. During the initial Brexit talks, the EU indicated that it would trade off market access against the degrees of integration which the UK would accept with the EU, comparing this to models of its relationships with other non-EU countries: Barnier’s even less sexy ‘escalator’.

This explains the nature of the UK/EU relationship, as eventually devised. First of all, though, the two sides agreed a withdrawal agreement which wound down the UK’s membership of the EU, retaining some continuing alignment with EU law, especially for the bitter ex-couple’s troubled child: Northern Ireland. That entailed continued jurisdiction for the CJEU and ongoing application of the direct effect and supremacy principles of EU law, rather than the more usual UK approach to international law, keeping it separate from the national legal order.

After the divorce, though, the parties agreed a very different type of treaty: the Trade and Cooperation Agreement (TCA), which co-exists alongside the withdrawal agreement. It explicitly eschewed any link with EU law, and therefore any CJEU jurisdiction (except as regards UK participation in EU research programmes), ruling out the ‘direct effect’ principle allowing individuals to invoke the treaty in courts to obtain rights – except as regards the social security provisions.

It proclaimed itself as the single framework for the UK/EU future relationship, with all further treaties between the two regarded as supplementary. But in fact it was itself several different treaties under a trenchcoat, containing within it a number of different rules on dispute settlement as well as separate termination and suspension rules for certain parts of the treaty.

Five years later, and a different UK government, believing there are economic benefits in closer ties – but political jeopardy in very close ties – with the EU, returned to the bottom of the Barnier escalator, hoping that closer ties could be negotiated without stepping on it. Ultimately this has proved possible, with the EU and UK willing to agree a complex compromise between both a number of substantive areas of interest and as regards the integration/market access trade-off in some particular fields.

The following blog post is an overview of the legal aspects of the reset deal – both substantively and in terms of the legal nature of what the two sides have agreed in principle.

Legal Framework

The reset arrangements are set out over three documents: a joint statement on broad foreign policy issues; a security and defence partnership; and a common understanding on development of the UK/EU relationship.  None of this is legally binding as such; but some of it is legally binding-adjacent, in the sense that it is linked to legally binding texts or indicates an intention to negotiate them. While operational cooperation can be important in practice, this blog post focusses on the issues with a link to legally binding measures.

Alongside these documents, the EU and UK have also agreed (as briefly mentioned in para 3 of the common understanding) to extend the current arrangements for fisheries and energy cooperation in the TCA. More precisely, according to the Commission’s Qs and As on the reset, the EU and UK have reached a ‘political agreement’ to extend the rules on fisheries in the TCA for 12 years, and to keep extending the rules on energy in the TCA annually. (Both sets of rules had been set to expire on 30 June 2026)

They have agreed to formalise this agreement to extend those rules ‘within one month’. It is not clear whether the parties think that this can be done via a decision of the Partnership Council set up by the TCA, or will need an amendment to that treaty by means of a Protocol, which would then need signature and ratification on each side. In light of the agreed speedy timeframe, we will soon find out the answer to that question.

Security and defence partnership

As the common understanding notes (para 6), agreeing the partnership potentially enables the UK and EU to agree on defence procurement within the context of the ‘SAFE’ Regulation on defence spending, under discussion at EU level (see the proposal here). The partnership text also refers to the UK’s participation in the EU’s military mobility project (approved under the previous UK government) and possible participation in other EU crisis management operations, along with association with training and the European Defence Academy. But it mainly provides for more frequent discussions between the EU and the UK in various security and defence fields.  

Common understanding

First of all, the much-discussed prospect of a ‘youth experience’ scheme is mentioned in para 13:

13. Therefore, the European Commission and the United Kingdom should work towards a balanced youth experience scheme on terms to be mutually agreed. The scheme should facilitate the participation of young people from the European Union and the United Kingdom in various activities, such as work, studies, au-pairing, volunteering, or simply travelling, for a limited period of time. It should provide a dedicated visa path and ensure that the overall number of participants is acceptable to both sides.

On the EU side, the Commission already proposed a mandate from the Member States (ie the Council) to negotiate a treaty to this end in April 2024; this proposal is public. The Council apparently agreed in principle on a mandate in December 2024, but does not seem to have formally adopted it; this might differ from the Commission proposal somewhat but we cannot be certain, as none of the Council texts are fully public at time of writing.

Looking at what the EU and UK have mutually agreed, the mobility will be limited in time, but the extent of this is not determined yet (the Commission had proposed perhaps three or four years). It will be on the basis of a visa (as the Commission had proposed). It appears that some form of limit on numbers will exist (the Commission had rejected this).

The agreement does not refer to a number of key issues, such as: the age range concerned (the Commission proposed 18-30); whether this will be a formal treaty, as the Commission proposed (which would need to be signed and ratified; on the EU side the Member States, via the Council, would have to agree to this, plus the consent of the European Parliament for conclusion); equal treatment in tuition fees (although not student loans), as the Commission proposed; family reunion (mentioned by the Commission); or dropping the UK’s healthcare surcharge (as the Commission proposed).

It should be noted that while some claim that ‘the UK asked for this first’, the Commission’s proposal makes clear that the UK had asked some Member States bilaterally for a youth exchange treaty; the EU preferred to respond by requesting a negotiation with the entire EU. There’s an obvious difference between separate arrangements with some Member States and a treaty open to nationals of all Member States.

If agreed, this would be the EU’s first youth exchange treaty (as distinct from whatever arrangements individual Member States have negotiated), and there’s no framework in EU law for such arrangements. Conversely, the UK has several youth exchange arrangements with non-EU countries.

Para 14 of the common understanding refers to Erasmus:

14. Furthermore, the United Kingdom and the European Commission should work towards the association of the United Kingdom to the European Union Erasmus+ programme. The specific terms of this association, including mutually agreed financial terms, should be determined as part of that process in order to ensure a fair balance as regards the contributions of and benefits to the United Kingdom. The association should be in accordance with the European Union Multiannual Financial Framework and the Trade and Cooperation Agreement.

(nb other non-EU countries are linked to Erasmus already). The reference to the TCA links the UK’s participation in Erasmus to the general rules for UK participation in EU programmes set out in the TCA already; Erasmus is not listed but Article 710 of the TCA gives the joint specialist committee set up by the agreement the power to amend the list of which programmes the UK participates in. So legally the two sides will presumably adopt something similar to the texts associating the UK with the EU research programmes; and like that negotiation, the negotiations will probably focus on the amount of the UK’s contribution. (Let’s not kid ourselves that either the EU or the UK is too high-minded to haggle over money).

Incidentally, as noted already, participation in EU programmes is the one part of the TCA which currently provides for jurisdiction of the CJEU to settle disputes (see Article 728, and my discussion of dispute settlement under the TCA); this will presumably apply to Erasmus too.

Next, there is a paragraph on touring musicians et al:

15. The European Commission and the United Kingdom recognise the value of travel and cultural and artistic exchanges, including the activities of touring artists. They will continue their efforts to support travel and cultural exchange.

Effectively this says nothing – the UK’s previous government having passed up an offer, when negotiating the TCA, to agree declarations similar to those attached to the EU’s visa waiver treaties with some other non-EU countries, which clarify that artists are not required to obtain visas during short visits to EU Member States, as they are not considered to be carrying out paid work.

On the issue of the use of e-gates, the common understanding says:

16. The United Kingdom and the European Commission will continue their exchanges on smooth border management for the benefit of their citizens, including the potential use of eGates where appropriate. They note that European Union citizens can use eGates in the United Kingdom and that there will be no legal barriers to eGate use for British Nationals traveling to and from European Union Member States after the introduction of the European Union Entry/Exit System.

Again, this is not really a commitment to anything. In practice, under the current EU border rules, which make no mention of e-gates, some Member States allow UK citizens to use e-gates before getting their passports stamped. Under the EU entry-exit system, supposedly to be applied later this year (although prior deadlines have been missed), e-gates are referred to explicitly; they will be used at least on some occasions to collect entry and exit data, with passport stamping becoming the exception. (EU legislators have just agreed on amendments to the law to roll out the system more gradually)

Moving on to economic issues, the common understanding refers to a possible agreement on the UK’s link to the EU electricity market:

18. The United Kingdom and the European Commission share the view that close cooperation on electricity is in the interest of both the European Union and the United Kingdom.

19. The European Commission and the United Kingdom should explore in detail the necessary parameters for the United Kingdom's possible participation in the European Union's internal electricity market, including participation in the European Union's trading platforms in all timeframes. Meanwhile, the current electricity trading arrangements will continue to apply.

20. Regarding its territorial scope, any agreement should be appropriately articulated with the provisions of the Windsor Framework. Any agreement should be based on a balance of rights and obligations and ensure a level playing field. In this context, it should define the relationship between the United Kingdom and European Union rules on the electricity market, as well as on State aid, the promotion of renewables and the protection of the environment, in so far as they relate to the electricity sector.

21. Accordingly, any agreement should include dynamic alignment with European Union rules where relevant, giving due regard to the United Kingdom's constitutional and parliamentary procedures and respect the role of the Court of Justice of the European Union within an arbitration-based dispute resolution mechanism, and an appropriate United Kingdom contribution to decision-shaping.

This appears to contemplate a future treaty, with a broad indication of its content – although of course the details would need to be renegotiated. Although the TCA already has rules on State aid and the environment, this text appears to consider that sui generis rules will be needed, including ‘dynamic alignment’ (ie the UK aligning with current and future EU law in this field) entailing a role for the CJEU, in which arbitrators (in the event of a dispute on EU law) will ask it questions about the interpretations of EU law.

This model is not used in the TCA – as noted above, the CJEU only has jurisdiction in the current TCA in the event of disputes about UK participation in EU programmes – but it is used in the withdrawal agreement (see discussion here), and in a number of EU treaties with non-EU countries. In practice, I don’t believe that any arbitration panel under an international agreement has ever asked the CJEU questions about EU law; and in some ways this would be less far-reaching than the CJEU’s role under the withdrawal agreement, given that some parts of that agreement (the Northern Ireland protocol; the financial settlement; citizens rights) provide for some additional jurisdiction for the Court.

A role for the UK in decision-making is likely to take the form of some kind of consultation similar to treaties with Norway et al (for instance, the EEA).

The detailed outline of a possible food standards agreement raises similar issues on dispute settlement, dynamic alignment, and a UK rule in decision-making, although it is slightly differently worded on these points:

23. The European Commission and the United Kingdom share the view that a functioning sanitary and phytosanitary area would address many of the issues raised in respect of the movement of agri-food products.

24. The United Kingdom and the European Commission should work towards establishing a Common Sanitary and Phytosanitary Area by way of a European Union-United Kingdom Sanitary and Phytosanitary Agreement (hereafter ‘SPS Agreement').

25. Regarding its territorial scope, the SPS Agreement should cover the European Union and the United Kingdom in respect of Great Britain (2). This would result in the vast majority of movements of animals, animal products, plants, and plant products between Great Britain and the European Union being undertaken without the certificates or controls that are currently required by the rules within the scope of the SPS Agreement for such movements. These same benefits would be extended to the movements between Great Britain and Northern Ireland, through the interplay of the Windsor Framework and the SPS Agreement, so long as the SPS Agreement is fully implemented. The continued application of the Windsor Framework would provide for Northern Ireland maintaining its privileged unique dual access to both the European Union Single Market and the United Kingdom internal market.

26. In terms of its material scope, the SPS Agreement should cover sanitary, phytosanitary, food safety and general consumer protection rules applicable to the production, distribution and consumption of agrifood products, the regulation of live animals and pesticides, the rules on organics as well as marketing standards applicable to certain sectors or products.

27. Within the scope defined above, the SPS Agreement should ensure the application of the same rules at all times by providing for timely dynamic alignment of the rules applicable to and in the United Kingdom acting in respect of Great Britain with all the relevant European Union rules, giving due regard to the United Kingdom's constitutional and parliamentary procedures; and where necessary to ensure the European Union's level of food, sanitary, and phytosanitary safety, through the immediate application of the relevant European Union rules.

28. The United Kingdom should be able to take targeted action to protect its biosecurity and public health, in the same way as Member States under European Union law. In addition, the SPS Agreement should include a short list of limited exceptions to dynamic alignment. An exception could only be agreed if: (i) it does not lead to lower standards as compared to European Union rules, (ii) it does not negatively affect European Union animals and goods
being placed on the market in the United Kingdom in respect of Great Britain, and (iii) it respects the principle that only animals and goods compliant with European Union rules move into the European Union.

29. The SPS Agreement should be subject to a dispute resolution mechanism with an independent arbitration panel that ensures the Court of Justice of the European Union is the ultimate authority for all questions of European Union law.

30. To ensure that it can put forward its view, the United Kingdom should be involved at an early stage and contribute appropriately for a country that is not a member of the European Union to the decision-shaping process of European Union legal acts in the fields covered by the obligation to dynamically align. The European Commission should consult the Government of the United Kingdom at an early stage of policy-making. These rights would not extend to participation in the work of the Council or its preparatory bodies.

31. The United Kingdom should have appropriate access to relevant European Union agencies, systems and databases in the areas covered by the SPS Agreement.

32. The SPS Agreement should provide for an appropriate financial contribution from the United Kingdom to support the relevant costs associated with the European Union's work in this policy area.

33. The SPS Agreement should be subject to a joint governance mechanism.

There is also a difference in territorial scope, in that a food standards agreement would not apply to Northern Ireland, the Windsor Framework/Northern Ireland Protocol addressing this issue already there. A UK financial contribution is expressly mentioned.

Again, the issues of costs, dynamic alignment, dispute settlement and decision-making are addressed as part of the section on emissions trading and the carbon border adjustment mechanism:

34. The European Commission and the United Kingdom share the view that a functioning link between carbon markets would address many of the issues raised in respect of trade and a level playing field and would give effect to Article 392(6) of the Trade and Cooperation Agreement. This link should not constrain the European Union and the United Kingdom from pursuing higher environmental ambition, consistent with their international obligations.

35. Therefore, the United Kingdom and the European Commission should work towards establishing a link between carbon markets by way of a European Union-United Kingdom agreement linking the United Kingdom Emission Trading Scheme (UK ETS) and the European Union Emission Trading System (EU ETS).

36. Regarding its territorial scope, this agreement should be appropriately articulated with the provisions of the Windsor Framework.

37. The agreement to link the UK ETS and EU ETS should create the conditions for goods originating in our jurisdictions to benefit from mutual exemptions from the respective European Union and United Kingdom Carbon Border Adjustment Mechanisms subject to compliance with the relevant provisions of European Union and United Kingdom legislation.

38. In terms of material scope, the agreement should cover all aspects of the functioning of an ETS link.

39. The sectors falling in the scope of the ETS linking agreement should be clearly defined to avoid risks of carbon leakage and competitive distortions. Among others, this scope should include the sectors of electricity generation, industrial heat generation (excluding the individual heating of houses), industry, domestic and international maritime transport and domestic and international aviation. The agreement should provide for a procedure to further expand the list of sectors to be covered by the linking agreement.

40. Within that scope, the agreement should ensure the dynamic alignment of the United Kingdom with the relevant European Union rules underpinning the functioning of the ETS link, giving due regard to the United Kingdom's constitutional and parliamentary procedures.

41. The United Kingdom cap and the United Kingdom reduction pathway will be guided by the United Kingdom's Climate Change Act obligations and Nationally Determined Contributions. They should be at least as ambitious as the European Union cap and the European Union reduction pathway.

42. The agreement should provide for an appropriate financial contribution from the United Kingdom to support the relevant costs associated with the European Union's work in this policy area.

43. The agreement should be subject to a dispute resolution mechanism with an independent arbitration panel that ensures the Court of Justice of the European Union is the ultimate authority for all questions of European Union law.

44. To ensure that it can put forward its view, the United Kingdom should be involved at an early stage and contribute appropriately for a country that is not a member of the European Union to the decision-shaping process of European Union legal acts in the fields covered by the obligation to dynamically align. The European Commission should consult the United Kingdom at an early stage of policy-making. These rights would not extend to participation in the work of the Council or its preparatory bodies.

45. The agreement should be subject to a joint governance mechanism.

There is rather less ambition as regards temporary entry and stay of businesspeople:

46. The United Kingdom and the European Commission will set up dedicated dialogues on the implementation of the Trade and Cooperation Agreement, as regards entry and temporary stay of natural persons for business purposes, including the sponsorship scheme, and the recognition of professional qualifications.

As for competition law cooperation, the common understanding refers to a recently agreed treaty, which was already foreseen in the TCA:

47. The European Commission and the United Kingdom recognise the mutual benefit of cooperation in competition enforcement. In this context and in implementing the level playing field provisions of the Trade and Cooperation Agreement, the European Commission and the United Kingdom welcome the successful conclusion of negotiations for a competition cooperation agreement between the European Union and the United Kingdom.

As it happens, the Commission has since proposed the signature and conclusion of this treaty.

Moving on to justice and home affairs, the common understanding encourages more use of the provisions in the TCA. On irregular migration, a hot topic on both sides, there are five paragraphs which effectively amount to very little:

57. The European Commission and the United Kingdom underline their commitment to deepen co-operation on challenges posed by irregular migration – including action to tackle people smuggling and to deepen information sharing to control and manage migration at our respective external borders – while remaining committed to ensuring international protection for those who need it.

Upstream Migration

58. The United Kingdom and the European Commission will deepen cooperation on the external dimension of migration across key geographies, as part of a whole of route approach. They recognise the benefit of increased mutual sharing of information, country expertise and analysis enabling better coordination. They will also work more closely in multilateral fora, processes and dialogues, including the European Union's Global Alliance to Counter Migrant Smuggling, and through exploring United Kingdom participation in the Khartoum and Rabat Processes.

Working together on practical solutions and returns

59. The European Commission and the United Kingdom should work together on practical and innovative approaches to reduce irregular migration. This includes work to share best practices and operational information on key issues like returns to third countries, and to explore possible deterrence mechanisms and other innovative solutions, in line with national, European Union and international law. This would be underpinned by closer cooperation across key networks such as the European Migration Network and agencies, including the European Union Agency for Asylum and Frontex, and respective United Kingdom authorities. They should work together to prevent irregular Channel crossings.

Bolstering United Kingdom and European Union border security including through law enforcement cooperation

60. The United Kingdom and the European Commission should bolster their operational and strategic cooperation to tackle the challenges posed by irregular migration along key routes and at external borders, including through their agencies, law enforcement and judicial cooperation, and information sharing capabilities. This includes through enhanced cooperation with Europol and its European Migrant Smuggling Centre.

Addressing challenges and abuses of visa policy

61. In addition to these enhanced efforts, the European Commission and the United Kingdom will explore closer cooperation to address challenges related to visa policy such as exchanging analysis on visa abuse by third country nationals.

It’s striking that the common understanding does not even mention the previous agreement for the UK to cooperate with Frontex, the EU border agency.

Finally, the common understanding limits its territorial scope in footnotes:

(1) : Unless otherwise provided, the territorial scope of the cooperation envisaged by this Common Understanding will be in line with the scope of Article 774 of the Trade and Cooperation Agreement.

(2) : In line with Article 520(5) of the Trade and Cooperation Agreement, the territorial scope of the SPS Agreement could be extended to cover the territories referred to in Article 774(2) of the Trade and Cooperation Agreement.

Article 774 TCA provides that in principle the TCA only applies to the UK (although note the specific references to Northern Ireland in the common understanding), while Article 774(2) provides for parts of it to apply to the Channel Islands and the Isle of Man. There is no mention of Gibraltar, which is excluded from the TCA by virtue of its Article 774(3); there is equally no mention of the separate treaty on Gibraltar that the EU and UK are negotiating. The Rock that dare not speak its name.

The overall legal relationship

As noted already, the TCA announced itself as the permanent template for the UK/EU relationship. Even though it had to co-exist with the more integrationist withdrawal agreement, the withdrawal agreement was the past and the TCA was the future, boldly going to a ‘dualist’ international law model separated from substantive EU law, its principles and its courts.

But now, the reset agreement makes little reference back to the TCA, besides the extension of the fisheries and energy provisions. Will treaties on youth mobility, and on food standards and emission standards/carbon border adjustment and the electricity market, take the form of ‘supplementary agreements’ to the TCA at all? Even if they do, they will not resemble the TCA legal framework that much – given that with the treaties on food standards, emission standards/carbon border adjustment and the electricity market, the current UK government has pragmatically accepted the integration/market access trade-off (as regards dynamic alignment and the CJEU) that the previous government rejected as ideologically unacceptable (under the TCA, that is).

And the question may also be asked whether these treaties, and the youth mobility treaty, will be subject to the principle of direct effect, ie invocability by individuals in courts, given that the social security rules in the TCA are already subject to an exception on this point.

Next, to what extent will all these treaties be connected? The extensions of parts of the TCA will remain connected to that treaty (with all its internal disconnections) of course, but will the other treaties be connected to the TCA – and/or each other? Inevitably this issue arises because some UK opposition parties have pledged to terminate the treaties concerned if they are elected to government; so the conditions for their termination will be relevant. This might also be relevant to negotiating these treaties – ie the UK (or EU) can’t have its pudding until it’s eaten its dinner.

Finally, it’s notable how many Rubicons have been crossed with this reset deal. As noted already, the UK now accepts the market access/integration trade-off. But the EU now accepts agreeing this trade-off with the UK in limited fields: the UK can have one foot several steps up the Barnier escalator, but the other one firmly on the ground. The EU has also accepted a Swiss-like complex legal relationship with the UK, having opposed it in principle for years. (In fact, the EU already conceded this point when agreeing the TCA; but that treaty hid its legal complexity better than the reset deal does). The UK has accepted an agreement with the EU as regards movement of (some) EU citizens; although it might claim this arrangement will simply resemble its youth mobility treaties with many other countries, the extent of that similarity will be dependent upon the details of the final deal. Above all, the EU, having accepted freer movement of some goods and demanded the freer movement of some people, can no longer lecture the UK on cherry-picking or cake-eating – what with all the crumbs and cherry juice smeared across the EU’s own mouth.

Pirates of the Mediterranean meet judges of the Kirchberg: the CJEU rules on Malta’s investor citizenship law

 

Steve Peers, Professor of Law, Royal Holloway University of London*

Photo credit: Aldo Ardetti, via Wikimedia Commons

*Thanks to Justin Borg-Barthet for corsair quote

**This blog post builds on research for the upcoming third edition of The EU Citizenship Directive: A Commentary, by Elspeth Guild, Steve Peers and Jonathan Tomkin (OUP, 2026)

 

Defending his country’s investor citizenship policy, a former Prime Minister of Malta once defined his homeland as a nation of ‘corsairs’, offering an explanation based on the legality (or not) of buccaneering behaviour:

“The difference between pirates and corsairs is that pirates do things illegally while corsairs have a license to do things that are on the brink of what is acceptable by the mainstream.”

Nevertheless, the Court of Justice of the EU has now ruled that the Maltese investor citizenship policy is a breach of EU law. So the Maltese government are pirates after all. Or are they? Is the real swashbuckling lawlessness here in fact the foray of the Luxembourg judges deep into the waters of Member States’ powers, to seize as treasure national competence which is plainly not theirs?

Background

According to the Declaration on nationality of a Member State, attached to the original TEU (the Maastricht Treaty):

The Conference declares that, wherever in the Treaty establishing the European Community reference is made to nationals of the Member States, the question whether an individual possesses the nationality of a Member State shall be settled solely by reference to the national law of the Member State concerned. Member States may declare, for information, who are to be considered their nationals for Community purposes by way of a declaration lodged with the Presidency and may amend any such declaration when necessary.

The point was reiterated in the ‘Edinburgh Decision’ of Member States a few months later:

The question whether an individual possesses the nationality of a Member State will be settled solely by reference to the national law of the Member State concerned.

Earlier in 1992, the CJEU gave its first ruling that fully addressed the issue: Micheletti. In this case, a national court asked the CJEU whether a dual citizen of a non-EU country and an EU Member State could rely upon EU free movement law, even though another Member State had doubts that his Member State nationality (acquired via his parents’ nationality) sufficed to entail a genuine link with that Member State, given that his previous residence was outside the EU.

The Court of Justice responded by referring to Member States’ power to define their nationality, with an ensuing obligation of mutual recognition by other Member States:

Under international law, it is for each Member State, having due regard to Community law, to lay down the conditions for the acquisition and loss of nationality. However, it is not permissible for the legislation of a Member State to restrict the effects of the grant of the nationality of another Member State by imposing an additional condition for recognition of that nationality with a view to the exercise of the fundamental freedoms provided for in the Treaty. [emphases added]

It followed that:

…it is not permissible to interpret Article 52 of the Treaty [now Article 49 TFEU] to the effect that, where a national of a Member State is also a national of a non-member country, the other Member States may make recognition of the status of Community national subject to a condition such as the habitual residence of the person concerned in the territory of the first Member State.

In the Court’s view, ‘this conclusion is reinforced by the fact that the consequence of allowing such a possibility would be that the class of persons to whom the Community rules on freedom of establishment were applied might vary from one Member State to another’.

Despite the reference to ‘international law’ in the context of a Member State granting or removing its own nationality, the Court implicitly rejected the approach of international law as regards judging whether another Member State’s nationality had to be recognised. In particular, the ICJ’s Nottebohm judgment had previously stated that countries could disregard another country’s nationality if there was no ‘genuine connection’ with that country. (The Advocate General’s opinion in Micheletti expressly argued that Nottebohm was not relevant).

Moreover, the throwaway reference to Member States’ power to define their nationality being subject to the proviso of ‘having due regard to Community law’ was eventually clarified. Starting with the 2010 judgment in Rottmann, the CJEU ruled that a Member State’s decision to deprive its citizens of its nationality was not unlimited, despite the declaration and decision of 1992, because that resulted in the loss of citizenship of the EU:

It is clear that the situation of a citizen of the Union who, like the applicant in the main proceedings, is faced with a decision withdrawing his naturalisation, adopted by the authorities of one Member State, and placing him, after he has lost the nationality of another Member State that he originally possessed, in a position capable of causing him to lose the status conferred by Article [20 TFEU] and the rights attaching thereto falls, by reason of its nature and its consequences, within the ambit of European Union law.

The Court qualified its prior judgment in Micheletti:

The proviso that due regard must be had to European Union law does not compromise the principle of international law previously recognised by the Court, and mentioned in paragraph 39 above, that the Member States have the power to lay down the conditions for the acquisition and loss of nationality, but rather enshrines the principle that, in respect of citizens of the Union, the exercise of that power, in so far as it affects the rights conferred and protected by the legal order of the Union, as is in particular the case of a decision withdrawing naturalisation such as that at issue in the main proceedings, is amenable to judicial review carried out in the light of European Union law.

Having established that deprivation of nationality was within the scope of EU law, the Court went on to say that there was no breach of EU law in doing so in this case, given the international measures that refer to the prospect of deprivation of citizenship in case of fraud (referring to the Council of Europe Convention on Nationality and the UN Convention on the Reduction of Statelessness, and also to the Universal Declaration of Human Rights as regards arbitrary deprivation of nationality).

The Rottmann judgment led inevitably to further case law, given the importance of deprivation of nationality to any individuals facing it: Tjebbes (discussed here); C-118/20; C-689/21; and C-684/22. Broadly speaking, this case law is generally deferential to Member States’ decisions to deprive citizens of their nationality, but subject to some limits, particularly a need for a legitimate interest (criminal offences, absence from the territory, dual citizenship) as well as the principle of proportionality and the guarantee of procedural rights. (Note that British citizens’ loss of EU citizenship due to Brexit was not subject to this case law, because that loss followed from the UK choosing no longer to be a Member State – as confirmed by the CJEU in 2022 (see discussion here))

On the other hand, the Court did not have the opportunity to clarify whether there were also limits imposed by EU law on Member States’ rules on acquisition of their nationality – until the Commission challenged the Maltese rules on investor citizenship, on the grounds that they amounted to selling the nationality of a Member State (and therefore EU citizenship) to people who lacked a sufficient link to that Member State.

The Commission v Malta judgment

The Court’s ruling starts by addressing the Maltese government’s argument that the constraints set by EU law on acquisition of Member States’ nationality are much less than the constraints on deprivation of nationality: they should ‘be limited to a finding of significant breaches of the values or objectives of the European Union, which are general and systematic in nature’. In the Court’s view ‘[t]here is nothing in the wording or the scheme of the Treaties to support the inference that their authors intended to lay down, as regards the grant of the nationality of a Member State, an exception to the obligation to comply with EU law’ only in such cases. Moreover, ‘such an exception cannot be accepted since it would amount to a limitation of the effects attaching to the primacy of EU law, which falls within the essential characteristics of EU law and, therefore, within the constitutional framework of the European Union’.

The Court then moves to the Commission’s argument that Malta had infringed Article 20 TFEU (the definition of EU citizenship, based on holding the nationality of a Member State) and Article 4(3) TEU (the EU law principle of sincere cooperation). The judgment points out that the Treaties aim to ensure free movement of people, in conjunction with the principle of mutual recognition. Next, the Court refers to the political participation elements of EU citizenship, by which ‘Union citizens participate directly in the democratic life of the European Union’; the EU’s ‘functioning is founded on representative democracy, which gives concrete expression to democracy as a value, which is, under Article 2 TEU, one of the values on which the European Union is founded’ (citing its recent judgment on EU citizenship and membership of political parties, which I discussed here). According to the Court:

It follows that the exercise by the Member States of their power to lay down the conditions for granting their nationality has consequences for the functioning of the European Union as a common legal order.

The Court also refers to the Treaty provisions on consular protection of EU citizens. Then:

It is in the light of those various rights that the Court has held that the provisions relating to citizenship of the Union are among the fundamental provisions of the Treaties which are part of the framework of a system that is specific to the European Union and which are structured in such a way as to contribute to the implementation of the process of integration that is the raison d’être of the European Union itself and thus form an integral part of its constitutional framework…

Next, the Court repeats its long-standing catchphrase that ‘Union citizenship constitutes the fundamental status of nationals of the Member States’, providing reasoning for that conclusion: ‘in the light both of the scope of the rights attaching to Union citizenship…and of the fact that that status derives automatically from the fact of being a national of a Member State’.

Therefore:

Union citizenship is thus one of the principal concrete expressions of the solidarity which forms the very basis of the process of integration…and which is an integral part of the identity of the European Union as a specific legal system, accepted by the Member States on a basis of reciprocity

And it follows that:

The exercise of the Member States’ power to lay down the conditions for granting the nationality of a Member State is not, therefore, in the same way as their power to lay down the conditions for loss of nationality, unlimited. Union citizenship is based on the common values contained in Article 2 TEU and on the mutual trust between the Member States as regards the fact that none of them is to exercise that power in a way that is manifestly incompatible with the very nature of Union citizenship.

The Court then reiterated (based on prior case law) that ‘the bedrock of the bond of nationality of a Member State is formed by the special relationship of solidarity and good faith between that State and its nationals and the reciprocity of rights and duties’, extending that to entail a principle underlying EU citizenship:

In the same vein, it is clear from the very wording of the first sentence of Article 20(2) TFEU that citizens of the Union are to enjoy the rights and be subject to the obligations laid down in the Treaties. In accordance with Article 20(1) TFEU, the special relationship of solidarity and good faith between each Member State and its nationals also forms the basis of the rights and obligations reserved to Union citizens by the Treaties.

When defining that ‘particular relationship of solidarity and good faith…the definition of the conditions for granting the nationality of a Member State does not fall within the competence of the European Union, but within that of each Member State, which has a broad discretion in the choice of the criteria to be applied, provided that those criteria are applied in compliance with EU law’.

But as the Court had already noted, that national discretion was not unlimited, but was subject to EU law. Crucially, then, the Court asserts that:

…a Member State manifestly disregards the requirement for such a special relationship of solidarity and good faith, characterised by the reciprocity of rights and duties between the Member State and its nationals, and thus breaks the mutual trust on which Union citizenship is based, in breach of Article 20 TFEU and the principle of sincere cooperation enshrined in Article 4(3) TEU, when it establishes and implements a naturalisation scheme based on a transactional procedure between that Member State and persons submitting an application under that programme, at the end of which the nationality of that Member State and, therefore, the status of Union citizen, is essentially granted in exchange for predetermined payments or investments.

A programme of that sort amounts to the commercialisation of the granting of the status of national of a Member State and, by extension, Union citizenship, which is incompatible with the conception of that fundamental status that stems from the Treaties.

The Court then recalls the mutual recognition principle from Micheletti, expressing its concern in that context:

Transactional naturalisation, which is granted in exchange for predetermined payments or investments, is not only contrary to the principle of sincere cooperation, but is also liable, by its nature, to call into question the mutual trust which underlies that requirement of recognition, since that trust relates to the premiss that the grant of the nationality of a Member State must be based on a special relationship of solidarity and good faith justifying the grant of rights resulting, in particular, from Union citizenship.

Diving into the specific rules of the Maltese investor citizenship scheme, the Court states that three of the five rules are essentially based on forms of cash payment. Although the fourth condition requires ‘legal residence in Malta for a period of 36 months’, the Court dismisses this as being, in effect, a Potemkin village: in practice, ‘that condition does not amount to a requirement of actual residence in the territory, since the physical presence of the applicant on that territory is required only when biometric data are collected in order to obtain the residence permit and to take the oath of allegiance’. This conclusion was reinforced by the opportunity to reduce the residence requirement to 12 months in return for further cash payments, and by comparison to the more stringent rules – entailing longer and genuine residence – which apply to the ‘ordinary’ naturalisation procedure of Malta. As for the final condition – a series of suitability checks in the public interest – the Court takes the view that they do not alter the essentially commercial nature of the sale of citizenship.

Finally, the Court dismisses Malta’s argument that its investor citizenship scheme ‘takes into account certain connecting factors on the part of the applicants’, in particular links via investments and possible future links after obtaining nationality. In the Court’s view: there was no evidence that the scheme ‘provides for a specific and concrete examination of the relevance of certain investments in order to establish the existence and extent of an applicant’s ties with the Republic of Malta, or to enable the development of such links with that Member State, distinct from’ examining whether an applicant meets the criteria; the prospect of deprivation of nationality after granting it was limited in scope; and the scheme was sold as conferring free movement rights in any Member State. Although Malta denied responsibility for how private parties sold the scheme, the Court points out that it had authorised them to act on its behalf – so (implicitly) had the power to prevent them from adopting these sales techniques, but failed to use that power. So, as a sales pitch, Malta was not really selling its nationality – it was selling citizenship of the European Union. And (implicitly again) Malta was not linking its nationality to its territory – it was linking it to being somewhere else in the EU instead.

Comments

I’ll start with the specific issues arising from the judgment (the impact on investor citizenship schemes) and move to the general (the impact on EU citizenship and the nature of the EU legal order more broadly).

Investor citizenship schemes

Several questions arise here. First of all, what happens to those who have already obtained citizenship under the Maltese scheme? Is Malta now obliged to yeet the rich?

Secondly, can Member States now refuse to recognise another Member State’s citizenship, if it has been obtained via means of an investor citizenship scheme that breaches EU law? Thirdly, would any system of investment citizenship by a Member State necessarily breach EU law?

Fortunately, immigration lawyer Simon Cox has already mused over these issues. On the first point, he points out that Malta did not ask the Court to limit the temporal effect of its ruling, so arguably it applies to citizenship already granted on the basis of the illegal scheme. Although, as he notes, Member States may not be keen to rescind citizenship already granted, other Member States have done so in similar circumstances (I would add that the CJEU case law on deprivation of nationality, mentioned above, would likely accept this on public interest grounds). But for those Member States reluctant to deprive investors of citizenship, it is possible that the CJEU may be asked to rule on this point – either if the Commission seizes the Court again, or if a Maltese court (perhaps seized by an NGO opposing the national scheme) asks the Court questions about the consequences of its judgment in that Member State.

But can another Member State now refuse to recognise the investor citizenship granted by Malta? The judgment touches on that point but very ambiguously, noting that a transactional citizenship rule is ‘liable, by its nature, to call into question the mutual trust’ underpinning EU citizenship, without answering the implied question of whether it is legal to do so. There’s a practical question of how to identify the Maltese citizens in question: presumably their passports are not literally golden. Perhaps immigration officials could profile those Maltese citizens who are flashing the biggest Rolexes and surrounded by the youngest women.

Assuming Member States do this, it should not be assumed, in light of the Court’s ambiguity, that it has overruled its long-standing judgment in Micheletti. Where the Court wishes to reverse its prior judgments, it does so explicitly (see, for instance, Metock: ‘It is true that the Court held [an absurd interpretation of free movement law in a previous judgment]….However, that conclusion must be reconsidered.’) In any event, in my view, the Micheletti judgment was and remains rightly decided on the mutual recognition point: the ICJ’s Nottebohm judgment was, with respect, poorly reasoned (paying insufficient account to the basic international law principle of state sovereignty), and it is hard to see how EU free movement law could function if Member States are second-guessing who is validly a national of another Member State. Above all, Member States had already made it clear (in the 1992 declaration and decision) that in the context of EU law, it was up to each of them – not other Member States – to determine who their nationals are.  

What about alternative versions of investor citizenship? Certainly the general tone of the judgment is one of disgust for commercialisation of EU citizenship in general. But as Simon Cox points out, the judgment also seems to suggest that a genuine residence requirement would overcome the Court’s revulsion. A less blatant sales pitch might help too. We can only speculate how long the super-rich would have to endure the inconvenience of mooring their yachts in Valetta for, to satisfy the Court – because it only told us that the current Maltese law was insufficient, without detailing which alternative would be legal.

EU citizenship and the EU legal order

This train already left the station with the Rottmann judgment 15 years ago, but it bears saying again, as the Court confirms that this case law applies to also the acquisition of Member State nationality: this whole line of judgments, with respect, has no basis in the wording of the Treaties or EU legislation. Not only is there no reference to any EU law constraints on Member States’ nationality law in the Treaties, the 1992 declaration and decision make absolutely clear that Member States interpret the Treaty to mean that nationality of a Member State is defined solely by national law.  

The Court’s specific rationales for rejecting Malta’s proposed high threshold for EU law constraints on acquisition of Member State nationality throw more flimsy fuel on this unconvincing fire. Yeah, of course the authors of the Treaties said nothing about an exception from EU law as regards acquisition of Member State nationality – because they said nothing about EU law applying to Member States’ nationality law in the first place. In fact, the 1992 declaration and decision clearly state that the authors of the Treaties believe EU law doesn’t apply to Member States’ nationality law at all. “We are very concerned about what the authors of the Treaties wrote” says the Court – while ignoring what the authors of the Treaties actually wrote on this very issue. This is simply judicial gaslighting.

As for the primacy of EU law, this is just confused. Even assuming one agrees with the Court’s conception of the primacy of EU law, it is a means of enforcement of EU law in the event of a conflict with national law. That primacy tells us nothing about what the scope or substance of EU law is – whether it applies at all to the acquisition of national citizenship, and if so what constraints it places upon those rules.

I have to say that if I read this reasoning on an EU law exam, the student would get nowhere near a 2:1.

Having said that, the judgment is what it is. We cannot avoid having to grapple with the Court’s explication of how it conceives of EU citizenship, and what that might mean more generally. And here, despite the weak foundations of its demand to set constraints upon national citizenship, the Court develops many interesting principles.  

The prior case law on the deprivation of Member States’ nationality was focussed on what the consequent loss of EU citizenship might mean for the individual concerned, particularly as regards free movement. But as regards acquisition of nationality, in this judgment at least the Court is not concerned about the impact on the individual. It is not even wholly focussed – as you might reasonably expect of the sale of nationality as a means to free movement – on the impact on other Member States. Rather it focusses in many respects on the impact for the European Union.

And what a European Union that is. As in the earlier case law, the Court mentions free movement aspects of citizenship, but it now adds references to consular protection and especially political rights. Coupled with the Court’s disdain for the sale of nationality, it seems that in this judgment, EU law is no longer founded on the internal market; it is now based on the internal marketplace of ideas. 

The Court does not mention the rather inconvenient fact that EU citizens face financial thresholds to exercise free movement rights: benefit recipients can move from Hamburg to Hanover, but not from Bucharest to Bremen. While EU free movement law is more generous than the usual immigration law for those of modest means, it is still more stringent than the rules applying to citizens of a Member State moving within it – on financial grounds. The Court is Captain Renaut in Casablanca: shocked, shocked to find that citizenship rights are dependent on money in here.

The judgment also links EU citizenship to democracy and the values of the EU, asserting an impact of Member States’ rules on acquisition of nationality upon ‘the functioning of the [EU] as a common legal order’, and ‘the identity of the [EU] as a specific legal system’. This stress on the legal nature of the EU reads like an attempt at fashioning constitutional patriotism by bloodless legal technocrats; and it is inevitably undermined by the profound weaknesses of the Court’s legal reasoning as regards the existence of EU constraints upon Member States’ nationality law.

But there’s a broader stress on non-legal values here too. The overall vibe is of a court preparing itself with arguments to defend the rule of law and other EU values – despite contrary developments in one country across the Atlantic and, more cogently, certain Member States. We shall soon see how this might play out in the pending case about EU values and Hungarian LGBT restrictions. And, circling back to Member State nationality, it is hard to see how the Hungarian government’s plan to suspend the Hungarian nationality of dual citizens who are insufficiently orgasmic about Orban could possibly be compatible with EU law in light of this new judgment.

One issue is notable by its absence from the Court’s reasoning: the argument that the rules on acquisition of nationality of a Member State are an aspect of national identity, which the Treaties explicitly require the EU to respect – even though Malta argued the point. It’s a strong argument in general terms, but its strength may differ in separate cases. There’s obviously a better case for (say) the grant of Irish nationality to (most of) those born in Northern Ireland as being intrinsically linked to the history and identity of Ireland, than there is for the souvenir shop citizenship of the Maltese law in this case. But the Court should nonetheless have answered the objection.

Conversely, the Court explicitly uses this judgment as a vehicle to express itself on the ‘identity’ of the European Union: as we have seen already, as a ‘specific legal system’. Presumably the specificity of this legal system explains – although again the Court avoids saying this directly – why the judgment implicitly rejects the Commission’s argument that Member States must base their nationality law upon a ‘genuine link’ with the country (echoing international law), and instead develops an EU-specific set of principles. (The break with international law is another point of divergence with the case law on deprivation of citizenship; perhaps it didn’t help that the Advocate-General’s opinion inaccurately claimed that the Council of Europe Convention has no rules on acquisition of nationality).   

Rather, the basis for the EU law rules on the acquisition of Member State nationality is the ‘special relationship of solidarity and good faith between each Member State and its nationals’; and the test is apparently whether a particular national rule is ‘manifestly incompatible with the very nature of Union citizenship’. Applying that test, ‘a Member State manifestly disregards’ this requirement when it sells its nationality on a commercial basis. Note that the Court does not say that this would be the only type of rule on acquiring nationality which is, in its view, incompatible with EU law.

What might this mean in other cases? Simon Cox’s blog post usefully points out some examples of other rules which might be challenged as a barrier to acquiring nationality (rather than, as in this case, making it too easy to obtain in some cases). He refers to high fees and education requirements; I would add that, by analogy with the case law on loss of citizenship, EU law may give applicants for nationality procedural rights too. And the link the Court makes with the values of the EU should, for example, preclude any practice (as in Trump’s USA) of refusing citizenship based on criticism of the actions of a foreign government – still less arresting the applicant at a citizenship interview with a view to deporting him.

Conclusion

This is the Alien v Predator of constitutional moments. I’m equally unimpressed by both the arrogant Eurofederalism of the Court and the fridge-magnet nationalism of the Maltese government. But what, in an ideal world, would be the best way forward?

There’s certainly a case, given the impact of the grant of nationality upon other Member States, of some form of coordination of nationality law within the EU. This would also be consistent with the logic of declaring the existence of citizenship of the European Union: why should the notion of citizenship be subject to no common rules at all?

But the development of such principles needs to be consistent with the rule of law, not the Court’s untenable interpretation of the 1992 decision and declaration. I doubt that there is much prospect of a Treaty amendment to confer competence upon the EU in this field: Orban may be many things, but he is not a turkey who would vote for Christmas. As things currently stand, principles could be agreed on the basis of soft law, or even treaties between Member States, on the basis of a ‘coalition of the willing’ if need be.

As I doubt that such a coalition will be easily forthcoming, I think we will be stuck with the development of legal principles in this field based on further elaboration of the Court’s judicial activism for the foreseeable future.  Time will tell whether the Court’s judgment is an anomaly – a reaction to a particularly egregious national policy in the context of broader allegations of corruption – or a key part of further case law on nationality law in particular and the EU legal order in general. 

To adapt the words of a famous on-screen pirate, look at the Court: they are the captain now.