No One becomes an EU Citizen for Gold: On Sovereignty, Union Values and the Limits of Passport Sale in Commission v. Malta

By Prof. Jean-Yves Carlier and Dr. Eleonora Frasca, Université de Louvain (UCLouvain), members of Equipe droits et migrations (EDEM)

This is a revised version of extracts from the yearly case law column “Libre circulation dans l’Union européenne”, published in French in the Journal de droit européen, no. 4, April 2026 (forthcoming).

Without constituting the final word, the judgment of the Court of Justice of the European Union (CJEU) in Commission v. Malta (Citizenship by investment) undoubtedly sets a benchmark in case law concerning the intersection between Member States’ nationality and EU citizenship. The ruling condemns what was referred to as the issuance of “golden” passports. In doing so, the CJEU uses constitutional language to affirm explicitly that Union citizenship “constitutes” the fundamental status of the nationals of the Member States (para 92). The issue was no longer to review the effects of a positive or negative conflict of nationality in light of the effectiveness of EU citizenship – as in most cases previously referred for a preliminary ruling spanning more than 30 years from Micheletti (1992) to Stadt Duisburg (Loss of German nationality) (2024). In this infringement procedure, the Court examined access itself to EU citizenship through access to nationality of a Member State. The Court held that Malta failed to fulfil its obligations under Article 20 TFEU (EU citizenship) and Article 4(3) TEU (principle of sincere cooperation), finding that the Maltese Citizenship Act “establishes a transactional naturalisation procedure in exchange for predetermined payments or investments and thus amounts to the commercialisation of the grant of the nationality of a Member State and, by extension, of Union citizenship” (operative part). The judgment was intended to draw a red line in the name of the Union’s values: EU citizenship is not for sale.

Our analysis proceeds in four steps: first, we analyse the Court’s constitutional framing of EU citizenship as a special relationship of solidarity and good faith. It is worth noting that “good faith” is rendered in French as loyauté, in Italian as lealtà, in Dutch as loyaliteit, in Spanish as lealtad, and in German as Loyalitätsverhältnis. Second, we assess whether the rights attached to citizenship genuinely function as vectors of EU integration. Third, we examine alternative doctrinal grounds for establishing Malta’s infringement – particularly the notion of abuse of rights. Fourth and last, we evaluate the short-, medium-, and long-term implications of the judgment for mutual recognition, national sovereignty and the future limits of judicial scrutiny over similar, but not equally worrying, Member States’ nationality policies.

1. Citizenship as a special Relationship of Solidarity and Good Faith?

From a substantive perspective, this judgment concerns free movement of persons only indirectly, insofar as it touches upon the very essence and definition of EU citizenship. From a spatial perspective, the vast majority of Member States, which neither practice nor contemplate this type of sale of nationality, are not concerned. However, the case raises important questions of principle, particularly regarding the possible role of EU law in relation to Member States’ retained competences. Nationality remains the expression of sovereign power par excellence. On the basis of fairly evident legal principles, it constitutes one of the main spheres for the exercise of national sovereignty, while at the same time serving as the gateway to EU citizenship.

The Grand Chamber’s findings against the Maltese law on naturalisation by investment are all the more noteworthy as the Court did not follow the Opinion by Advocate General Collins. This is despite the fact that the Advocate General proceeded from the same premise as the Court: “The exercise of a Member State’s sovereign prerogative to grant or to withdraw citizenship is not unlimited and both EU law and international law may, in principle, constrain its exercise” (para 49 of the Opinion and para 81 of the judgment). However, the Advocate General considered that “one cannot equate a withdrawal of a Member State’s nationality with its acquisition” (para 52), thus implicitly accepting that EU law’s review of the loss of an already acquired EU citizenship might be more extensive than the review of access to EU citizenship through the acquisition of nationality.

Rejecting any exemption from review of compliance with EU law – expressed even more clearly by Malta –, the Court held that “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”, explicitly situating the constitutional scope of its assertion by referring to Opinion 2/13 on the Union’s accession to the ECHR (para 83). Accordingly, from the perspective of a possible infringement of Article 20 TFEU, the CJEU meticulously recalls all the rights associated with citizenship, ranging from free movement to political rights and diplomatic protection (paras 85 to 90). It specifies that “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” (para 91, emphasis added).

The CJEU further relies on the Grzelczyk formula, observing that “the Court has repeatedly held that Union citizenship constitutes the fundamental status of a national of a Member State” (para 92). Linking this to the principle of sincere cooperation and the possible infringement of Article 4(3) TEU, the Court notes that “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 manner that is manifestly incompatible with the very nature of Union citizenship” (para 95).

Finally, the Court states 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” (para 96). The conclusion follows: “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” which “amounts to the commercialisation of the granting of the status of national of a Member State and, by extension, of Union citizenship” (paras 99 and 100).

The analysis of the content of this CJEU ruling can be summarised in a single question. Is this what has been demonstrated or simply what had to be done? Most scholars acknowledge that the Court did, in the operative part of its judgment, what had to be done (among many, J. Coudron, S. Peers, A. Iliopoulou-Penott, K. Lamprinoudis and F. Jault-Seseke). The sale of nationality in exchange for nothing more than mere cash was simply shocking. It had to be condemned. If the outcome seems justifiable, the Court’s reasoning employed, for its part, has been criticized – sometimes sharply (among many, D. Kochenov, G. Athanasiou and J. Heymann). Many commentators consider that the Court’s reasoning leaves something to be desired, either because of its legal weakness or, paradoxically, because of its excessive constitutionalism, allegedly encroaching on national sovereignty (among many, S. Poli). It is true that the reasoning may come as a surprise on several points. Quantitatively, it is scarcely developed: while 78 paragraphs are devoted to the statement of facts and the parties’ arguments, only half as many concern the Court’s assessment. Qualitatively, various arguments, formulations, or references fail to persuade.

2. Rights of EU citizens as proxy for Integration in the Field of Citizenship?

One may accept the fundamental objective of integration within the constitutional perspective adopted by the Court. As the editorial of an issue of the European Law Review recalls, “[w]e have long been aware of the Court’s interpretative maxim in dubio, pro integratione”.[1] However, the mere enumeration of the rights of citizens does not in itself demonstrate the existence of an integration process. Something largely overlooked in the critiques is that most of these rights are also granted to certain categories of third-country nationals. This is the case for freedom of movement, which is possible not only to third-country nationals who are family members of citizens, but also to long-term residents. The same applies to the right of petition and the right to apply to the Ombudsman, both of which are open to any resident within the territory of the Union. Even the right to vote and to stand as a candidate, including in elections to the European Parliament, had been regarded by the Court – at least under the law as it stood at the time – as not precluding “Member States from granting that right to vote and to stand as candidate to certain persons who have close links to them, other than their own nationals or citizens of the Union resident in their territory” (Spain v. United Kingdom, para 78). The Court refrains from citing this case law about Gibraltar as well as the case law about Netherlands’ overseas territories of Antilles and Aruba (Eman and Sevinger) which, conversely, allowed citizens to be excluded from voting rights if the criterion was not discriminatory. That leaves diplomatic protection, to which the Court also refers and which, in fact, seems to be the only right still exclusively reserved for EU citizens. However, it must be acknowledged that, in international law, such protection granted by another State – one closely connected to that of the national – though not strictly a right, nevertheless results from consistent practice.

By contrast, the Court pays little attention to the rights from which many categories of citizens benefit only minimally, if at all – particularly in the field of free movement: namely, unconditional access to social assistance for citizens who lack sufficient means of subsistence. Consequently, grounding its reasoning in the Grzelczyk formula lacks realism, especially when it converts, without further explanation, the idea that citizenship is “destined to be” the fundamental status (Grzelczyk, para 31) into the assertion that citizenship “constitutes” the fundamental status (Commission v. Malta, para 92). Of course, this is not the first time, as some have written, that the Court has affirmed that European citizenship ‘constitutes’ a fundamental status (see Stadt Duisburg (Loss of German nationality), para 42 and Udlændinge- og Integrationsministeriet (Loss of Danish nationality), para 38). Nevertheless, this evolution in vocabulary is not sufficient to justify the foundation of European citizenship, derived from nationality, in “a special relationship of solidarity and good faith”.

Likewise, illustrating this special relationship of solidarity and good faith by referring to cases concerning the free movement of workers, such as Commission v Belgium or Lawrie-Blum, is somewhat surprising. Beyond the principle of free movement, those cases concern an exception: the right of States to reserve access to the employment in the public service for their own nationals (Art. 45, paragraph 4, TFEU). They deal precisely with situations in which, once again, the EU citizen may be excluded from certain rights that are exclusively reserved for nationals. In other words, however strictly the “employment in the public service” exception is interpreted, this very definition – based on solidarity and good faith – reveals not continuity but a rupture of rights between nationality and citizenship.

Criticism, however, is easy. The Court had to condemn this absolute commercialisation of EU citizenship. And if it was to do so, why not by emphasising a conception of citizenship that is more civic than market-based? To observe that the Court substitutes an economically grounded, freedom-of-movement-based citizenship with a civic conception rooted in fundamental rights – is this not to acknowledge that it is realigning with the foundational logic of the Ruiz Zambrano judgment (see, on this point: A. Iliopoulou-Penott)? Nevertheless, could the Court have relied on any alternative, less sovereignty-intrusive ground for condemnation of a Member State?

3. Where there any Alternative Grounds for establishing Malta’s Infringement?

It is well known that, “it is for each State to determine under its own law who are its nationals” according to the very wording of Article 1 of the 1930 Hague Convention. What is often forgotten is the remainder of the provision, which specifies that “this law shall be recognized by other States in so far as it is consistent with international conventions, international custom, and the principles of law generally recognized with regard to nationality”. This led the International Court of Justice (ICJ), in the Nottebohm case, to hold that nationality granted by Liechtenstein to Mr. Nottebohm – acquired through payment in order to avoid the seizure of his property by Guatemala as a German enemy national during wartime – was not effective, in the absence of genuine links, and therefore need not be recognized. In doing so, the ICJ defined nationality as “a legal bond having as its basis a social fact of attachment, a genuine connection of existence, interests and sentiments, together with reciprocal rights and duties”. In truth, in Commission v. Malta, the CJEU does not explicitly rely on the concept of “genuine connection”. In the words of Jules Lepoutre which we take the liberty to translate: “the Grand Chamber engages in Nottebohm without saying so, avoiding explicitly writing that it is assessing the ‘genuine connection’”. Unlike the Commission and Advocate General Collins, the Court refrains from citing the Nottebohm judgment – something of a scarecrow embodying a notion of effective nationality that, in the Micheletti case, Advocate General Tesauro wrongly described as “a concept harking back to the romantic period of international relations” (para 5).

Under international law, the consequence of the notion of effective nationality would be to allow, at most, another State to refuse to recognize the effects of a nationality lacking any genuine connection – not to challenge the legal fact that the person in question possesses that nationality. Yet this is precisely where the difficulty lies for the CJEU. Within the framework of the European Union and mutual trust, such a refusal of recognition – left to the discretion of the host State – would be problematic and contrary to the earlier Micheletti case law, which the Court invokes to recall that “Member States are required to recognise the effects of the granting to a person, by another Member State, of the latter’s nationality with a view to the exercise of the rights and freedoms arising from EU law” (para 101, emphasis added). However, such transactional naturalisations are liable “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 the rights resulting, in particular, from Union citizenship” (ibid.). Unable to leave to each Member State the discretion to recognise – or not – the nationality granted by another Member State on whatever criteria, the Court could only condemn or absolve. Unable to absolve, it condemned. But could the Court condemn on another basis?

If one accepts the “shocking” nature of these sales of nationality, the analysis might naturally shift to the concept of abuse of rights. This notion is closely related to that of fraud, the underlying concern in Nottebohm. It was also invoked in Chen, though ultimately dismissed (see judgment, para 34ff and Advocate General Tizzano Opinion, paras 109, 116 and 122). Fraud is often considered – if not always upheld – when assessing the conduct of individuals or legal entities, for example in tax or company law. In the case of the Maltese golden passports, however, the potential fraud would lie less with the individuals applying for (and obtaining) nationality than with an abuse of rights attributable to the State.

Establishing an abuse of rights would not be straightforward, since both its objective and subjective elements must be satisfied: the apparent formal compliance with rules whose underlying purpose is frustrated and the intention artificially to construct the conditions required to obtain an advantage. Is this path truly so arduous? The Court’s repeatedly underscores the considerable financial benefits that these sales of nationality accrue to the Republic of Malta, despite the absence of any other link with the objective of national belonging? Moreover, Malta’s own promotional campaigns for advertising these naturalizations emphasised not integration into the Maltese polity, but the advantages attached to the possession of EU citizenship…

4. On the Effects of Commission v. Malta

Did these elements not allow for a serious consideration of a finding of abuse of rights on the part of the Maltese State? The advantage of such an approach would have been to encroach less upon national sovereignty in matters of nationality attribution, while more clearly emphasizing that what is at issue is an exception to a State’s exercise of its sovereignty – limited to cases where that sovereignty is formally used for a purpose other than that of establishing a connection with the first of the three constituent elements of the nation-state: the people, the territory, and organized political authority. As the European Group of Private International Law (EGPIL) notes in its Guidelines on the influence of Union law on State nationality rights, approved at the 2025 Ljubljana meeting, after the ruling in question, “when granting their nationality, Member States shall ensure compliance with the general principle that Union law may not be invoked upon for abusive or fraudulent purposes”. Regardless of the reasoning employed by the CJEU, once the infringement has been established, we can consider its possible effects in the short, medium, and long term.

4.1 On Mutual Recognition

In the short term, there is no obligation to withdraw the Maltese nationalities that have already been granted. However, the question arises as to whether these nationalities must be recognized by the other Member States. How do the two “M” judgments intersect? Is the Malta ruling consistent with Micheletti? Some will argue that other Member States could rely on the infringement judgment to refuse recognition of the effects attached to that nationality – such as free movement – on the basis of a form of European public policy, although abuse of rights would arguably have offered a more solid doctrinal basis. Others will take the opposite view and maintain, conversely, that the effects of previously conferred nationality cannot be called into question precisely because the Court grounded its infringement’s findings in the Micheletti line of case law and in the principle of mutual trust. Both positions are defensible. Nevertheless, given both the Court’s reaffirmation of the Micheletti case law – according to which Member States “are required” to recognise the effects of a nationality granted by another Member State (para 101, cited above) – as well as the legal effects attaching to an infringement judgment, the second line of reasoning appears more persuasive.

4.2 Graded Nationality as a Sovereignty-Preserving Alternative to Infringement

In the medium term, those effects consist precisely in Malta’s obligation to amend the legislation found to be in breach of EU law. Failing that, it could, following a second action, be ordered to pay substantial penalty payments, comparable to those imposed on Hungary for failure to comply with obligations concerning the reception of asylum seekers. In the Maltese context as well, a finding of infringement grounded in abuse of rights would have opened the way to a more calibrated solution. The objective would not have been to grant absolution coupled with token penance – an outcome likely to invite Malta’s indifference – but rather to condemn the practice while indicating lawful avenues for reducing abuse.

One such avenue, indirect yet respectful of national sovereignty, lies in the technique of differentiated or graded nationalities or naturalization, whereby access to rights varies according to the status conferred. In principle, nothing would preclude the creation of a “national nationality” that does not entail access to EU citizenship. Admittedly, given the own wording of Article 20 TFEU which automatically links nationality to Union citizenship, another term would be required, akin to the category of “overseas citizens” once recognized by the United Kingdom. In Kaur, the CJEU itself accepted that this specific form of British citizenship did not confer a right of residence and, consequently, free movement rights to those belonging to the category. After reiterating the Micheletti formula – according to which national sovereignty falls within national competence subject to compliance with EU law – the Court added that “on the basis of this principle of customary international law, the United Kingdom has […] defined several categories of British citizens whom it has recognized as having rights which differ according to the nature of the ties connecting them with the United Kingdom” (Kaur, para 20). In light of the greater or lesser closeness of those ties, the degree of connection between the individual and the State thus becomes the criterion for allocating rights, including those attached to EU citizenship. And it is in this regard that, upon its accession to the EEC, the United Kingdom appended a declaration specifying which categories of its nationals were to be regarded as EU citizens.

Moreover, as indicated supra, in Spain v. United Kingdom (on Gibraltar), the CJEU upheld the extension of voting rights in European Parliament elections to those same overseas citizens treated as third-country nationals but “expressing a specific link with the territory” (para 79). It follows that, even absent a formal declaration and even in the context of an infringement not expressly framed as abuse of rights, Malta could adjust its legislation by modulating its categories of nationality. Such an approach would preserve national sovereignty while safeguarding the Union’s constitutional values. Over time, however, few applicants would seek to obtain a nationality that does not give access to EU citizenship. As a result, interest in Malta’s public finances would be likely to decline significantly.

4.3 Limits of Judicial Scrutiny over Nationality Policies and EU Citizenship

In the long term, the question arises as to the extent to which EU law may constraint Member States’ nationality policies where those policies are perceived as overly expensive, and by extension, as broadening access to EU citizenship rights. Critics have referred to the risk that certain modes of acquiring nationality might not satisfy the criterion of “special relationship of solidarity and good faith” articulated by the Court. One may think of naturalisations for merit (France), acquisition based on historical blood ties (Italy, see B. Nascimbene; C. delli Carri), access by belonging to a particular ethnic group (Hungary, see S. Poli and L. Lonardo) or to a defined territory (Ireland).

Such concerns, however, tend to overlook two important distinctions. First, it is necessary to differentiate between attribution of nationality at birth – whether iure sanguinis or iure soli – and naturalisation. The former operates automatically by virtue of law and statutory criteria, whereas the latter follows an individual decision taken in response to an application. Second, the various hypotheses of attribution or granting of nationality mentioned above, though politically sensitive, each reflect elements of a relationship of solidarity and good faith, albeit in different forms and degrees, individually or collectively. For example, the Republic of Ireland’s attribution of nationality to persons born on the island of Ireland, including in Northern Ireland, is not a case of naturalisation but of iure soli attribution. Furthermore, it embodies an historical understanding of solidarity and good faith with a particular people. Whatever one’s political assessment of the Irish approach, it is not legally comparable to the commodification of nationality (ius pecuniae) nor can it readily be characterised as an abuse of rights – an interpretation consistent with the reasoning in Chen.

What remains to be considered are other investment-based naturalisation schemes (Portugal, Spain). In truth, the Court’s close scrutiny of the Maltese legislation, in our opinion, considerably narrows the scope of sanctionable situations. The judgement only targets cases involving the outright commercialisation of nationality, without any other link. The CJEU thus emphasised that the Maltese legislation imposed no requirement of effective residence, 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” (para 106). As noted by EGPIL in its aforementioned Guidelines, the Malta judgement “did not introduce a review of effectiveness in the strict sense, but simply prohibited the commercialisation of nationality when it was not accompanied by any other verification of ties”. It is therefore difficult to identify which Member State could, on that basis, fall within the Commission’s crosshairs. Accordingly, we do not endorse the view that the Court created a “dangerous precedent, one which carries exclusionary potential and which can jeopardise the legal status and rights of an undetermined number of Europeans” by opening “the door for itself to decide who may become the citizen of a Member State, substituting the EU’s judgment for that of the Member States when acting in the field of their competence” (see D. Kochenov and G. Íñiguez).

Drawing directly from the judgment as well as the Advocate General’s own observations in his Opinion (despite different conclusions), the clarifications set out above indicate otherwise. In his opinion, Advocate General Colins had clearly stated that even if the infringement procedure were successful, the consequence would not be, either for Malta or for any other State, to “impugn the legality of that Member State’s entire legislative framework on naturalisation” (Opinion, para 36). The infringement action brought by the Commission does not constitute “an effort by that institution, with the Court’s direct or indirect assistance, to legislate in the field of EU citizenship” (ibid.). Nothing therefore prevents us from reaffirming that in matters of nationality “the principle of State competence remains firmly anchored” (echoing Paul Lagarde’s remarks on the Rottmann judgment in our own translation). Sovereignty, in matters of nationality as in others, must accommodate the realities of supranational constructions and, in particular, the requirements of EU law. Yet adaptation does not necessarily entail subordination where it is exercised with discernment and reciprocity, in a spirit of mutual respect for competences.

In the longer term, another possibility remains: an intervention by the EU legislator to establish a form of Union citizenship detached from the nationality of Member States, for example on the basis of regular residence exceeding five years. We have long been in favour of this approach (see J.-Y. Carlier). However, this would more accurately constitute a process of “citizenisation” or “ EU naturalisation” rather than the attribution of nationality and citizenship at birth. In essence, it would amount to converting long-term resident status into Union citizenship (see L. Moccia). Nevertheless, defining Union citizenship independently of nationality appears politically unrealistic, particularly in these times of strong national identities. We believe that a more appropriate solution lies in striking a balanced accommodation through mutual adaptation between the Union and the Member States, and between citizenship and differentiated or graduated forms of national membership.

Conclusion: Shift rather than Rupture

The Court’s decision can be endorsed. This ruling condemning golden passports should not be seen as a breaking point in the link between nationality and Union citizenship. It represents continuity with previous case law in framing sovereignty in matters of nationality. This framing is marginal, highly exceptional, yet certain insofar as it is rooted in the values underpinning the Union which also underlie the democratic constitutions of the Member States. In that sense, the judgement constitutes a measured development rather than a turning point: a subtle shift in emphasis rather than a rupture. It could remain isolated unless further intervention proves necessary. It signals continuity rather than transformation.

[1] “Of ends and means”, E.L. Rev., 2025, 50(4), pp. 381-382.