Union citizens’ rights against their own Member State after Brexit

The treatment by the United Kingdom of Union citizens remaining on its territory after Brexit and conversely that of UK nationals by EU27 Member States on theirs has given rise to much discussion and analysis. By contrast, there has been comparatively little systematic and detailed exploration of the question of the impact of Brexit on the exercise of Union citizens’ rights against their own Member State. It is an issue which is for the most part ignored in the current Withdrawal Agreement. The purpose of this article is to show that this blind spot opens up a potential gap in legal protection of the rights of Union citizens, which is likely to remain regardless of the outcome of the Brexit negotiations and whether a withdrawal agreement is concluded or not. The paper discusses the extent to which the adversarial nature of the withdrawal process has contributed to this failure to address this issue and the ways in which courts could step in to provide the legal protection that political processes were unable to deliver.


Introduction
The consequences of the withdrawal of the United Kingdom from the European Union on the rights of Union citizens is clearly an important aspect of the Brexit process. Indeed, right from the start, safeguarding those rights was identified in the first set of European council Guidelines for Brexit negotiations as the 'first priority' for these negotiations. 'Effective, enforceable, non-discriminatory and comprehensive' guarantees were to be provided to enable Union citizens to 'exercise their rights through smooth and simple administrative procedures'. 1 It also seems to have been on the whole a somewhat easier and less controversial aspect of the Brexit negotiations. While discussions on the other two main topics addressed in the initial phase of the process, namely settlement of the UK's financial obligations and avoidance of a physical border on the island of Ireland were protracted, there was a greater element of consensus between the negotiators on how to address the question of citizens' rights after Brexit, notwithstanding the persistence of some divergent perspectives on specific aspects right up to November 2018, when agreement was reached on a Withdrawal Agreement. 2 While negotiations were re-opened in Autumn 2019 following the coming into office of a new UK government in the summer, the revised text of the Withdrawal Agreement agreed in November 2019 3 has not amended Part Two of the Agreement, 4 in which the provisions devoted to citizens' rights are contained. Both sides of the negotiations would seem therefore to regard the consensus previously achieved on citizens' rights as acceptable.
Yet the Withdrawal Agreement would potentially leave a significant gap in the protection of citizens' rights after Brexit. Union citizenship endows citizens and their families with rights not only vis-à-vis other Member States but also with respect to their own Member State. While not perfect, the Withdrawal Agreement has much to say about the former. In relation to the latter, however, the Agreement is largely silent. The adversarial nature of the Article 50 TEU process has tended to obscure this dimension of Union citizenship. The focus has been on the UK protecting the rights of their nationals residing in the EU27 Member States and on EU institutions protecting the rights of EU27 citizens residing in the UK without much consideration being given as to how the rights of individuals vis-à-vis public authorities within their own camp, so to speak, would be affected.
The Withdrawal Agreement has not yet been ratified and it is still unclear at time of writing 5 whether it will be and, if so, when. 6 Whatever happens on this front is unlikely to make much difference to the present enquiry. If a withdrawal agreement -whether the current one or a yet further revised one -is eventually ratified by both sides, it would in all likelihood contain identical or very similar provisions on citizens' rights to those in the currently agreed text. A fortiori, if the UK were to leave the EU without a withdrawal agreement, citizens' rights would be even less protected.
This article aims to assess the extent of the gap in the protection of citizens' rights after Brexit resulting from the failure to sufficiently consider the implications of Brexit of the relationship between nationals and their own Member States. It is structured as follows. In the first place, the Withdrawal Agreement is not perfectly aligned 11 with the notion of family member in Directive 2004/38 on the free movement rights of Union citizens and their families (the Citizens' Rights Directive) 12 and family reunification will be more difficult in some circumstances where the family is not yet fully constituted and/or residing in the host State by the end of the transition period. In the vast majority of cases, however, the rights that family members would have enjoyed in the host State under Directive 2004/38 are grand-fathered by the Withdrawal Agreement. 13 Setting aside the specific provisions on social security coordination discussed below, two aspects of the personal scope of the citizens' rights provisions constitute a severe limitation on the possibility to invoke the benefits of the Agreement against one's own Member State: the requirement of continuous residence and the impossibility for individuals who have not exercised, or ceased to exercise, free movement of rights to invoke the agreement.

Continuity of residence requirement
While the grand-fathering of residence rights in the Withdrawal Agreement is in principle a lifelong one, 14 it is nonetheless subject to the condition that the individuals concerned continue to reside in the host State. 15 If an EU27 national decides to leave the UK or a UK national decides to leave the EU27 Member State in which they reside, the Withdrawal Agreement will cease to apply to them. 16 The problem with this is that returning home is a critical point in terms of being able to invoke union citizenship rights against one's own Member State. It is at this point that, for instance, a Union citizen is particularly likely to need to invoke a right of residence for their family members 11. The main differences are: (i) more restrictive conditions apply to family members (including as of yet unborn children) who were not residing in the host Member State at the end of the transition period and seek to join the Union citizen afterwards; (ii) persons who belong to the wider circle of family members as defined in Article 3(2) of the Citizens' Admittedly, in case of temporary visit to the home State, or elsewhere for that matter, the inapplicability of the citizens' rights provisions is not immediate. The provisions of the Citizens' Rights Directive on continuity of residence for the purpose of acquisition of permanent residence under the Directive are transposed to the question of continuity of residence for the purpose of applicability of the citizens' rights provisions in the Withdrawal Agreement. 19 An individual will therefore be able to leave the host State for up to six months per year or once for up to 12 months in case of an 'important reason' 20 without jeopardising the continued application of those provisions. In addition, if the individual has acquired a right of permanent residence in the host State, it would take an absence of five consecutive years from the host State for that permanent right of residence to be lost. 21 Nonetheless, in case of definitive, or at least prolonged, return, the individual will eventually fall out of the scope of the Withdrawal Agreement.

Non-applicability to non-movers and early returnees
The wording of Article 10 would also imply that individuals who have never left their home State or who have returned to their home State before the end of the transition period fall outside the scope of the Withdrawal Agreement. 22 In relation to the latter, it could mean, for instance, that a professional recently returning to the UK from a EU27 home State, or vice-versa, might conceivably lose after the end of the transition period the right they would have had under Directive 2005/36 23 to recognition in their home State of a professional qualification acquired in the host State and therefore the right to exercise their chosen occupation. It would also mean that the protection of the right of residence in the home State afforded by the Surinder Singh case law to family members of a Union national returning home 24 could come at an end once the transition period is over.
The non-applicability of the agreement to individuals who have never exercised rights to free movement between the UK and an EU27 Member State is more understandable. In the absence of any exercise of free movement rights, the situation will as a rule be regarded as a wholly internal situation falling outside the scope of EU law on citizens' rights. 25 There is however an exception to this under the Ruiz Zambrano 26 case law, in which the Court found that a third country national may have a right of residence in a Member State under Union law where refusing such a right of residence would result in a Union citizen who is dependent on that third country national being 17 forced to leave the territory of the European Union as a whole and therefore deprived of the 'genuine enjoyment of the substance of the rights attaching to the status of European Union citizen'. 27 Rights of residence under the Zambrano case law are normally exercised in the State of which the dependent Union citizen is a national. 28 Brexit would not in itself affect the rights of Zambrano carers of EU27 nationals. It would, for instance, still be possible for the Colombian carers of Belgian children to rely on the Zambrano case law to establish a right of residence in Belgium so as to ensure that their children are not deprived of the 'genuine enjoyment of the substance of the rights' attaching to their status as Union citizens. This, however, would no longer be true of the Zambrano carers of UK nationals since those UK nationals would no longer be Union citizens and therefore unable to invoke the need to protect the 'genuine enjoyment' of their rights as Union citizens. Leaving the issue of Zambrano carers of UK nationals to be determined in accordance with domestic UK law opens up a gap in the legal protection provided by the Agreement even though it remains open to national law to provide the protection that the Agreement does not offer. 29

B. Rights protected in the withdrawal agreement
Even if individuals were able to bring themselves within the personal scope of the Withdrawal Agreement, there remains the question of what rights they would be able to invoke.
Citizens' rights protected by the Withdrawal Agreement are enumerated in Title II of Part Two of the Agreement. 30 Title II is divided into three chapters devoted respectively to rights of residence, to rights of workers and self-employed persons and to professional qualifications. Title III is not expressed in terms of rights but its contents, social security coordination, nonetheless establishes important entitlements for individuals after Brexit. This section will loosely follow the structure used in the Withdrawal Agreement by distinguishing between migration rights, rights related to the exercise of an economic activity and rights related to social assistance and social security.

Migration rights
By migration rights, I mean rights to free movement stricto sensu, that is to say the right to cross a border, whether to leave or enter the territory of a Member State as well as right to stay within a Member State after having crossed the border, viz rights of residence.
Right of exit. Article 4 of the Citizens' Rights Directive expressly recognises the right to leave a Member State, including the Member of which one is a national, to travel to the territory of another 27. Ibid., para. 42-44. 28. See Case C-86/12 Alopka, EU:C:2013:645, para. 32-35. 29. Such protection is in fact provided by the UK Immigration Rules: See, UK Government, 'Appendix EU: EU, other EEA and Swiss Citizens and family members', UK Government (2019), https://www.gov.uk/guidance/immigration-rules/ immigration-rules-appendix-eu. 30. This title is curiously entitled 'Rights and obligations' even though the only obligations it contains seem to be conditions to be fulfilled to be able to invoke the rights rather than self-standing obligations.
Member State. In so doing, Article 4 echoes the recognition of the right to leave a country, including one's own, found in multiple human rights instruments. 31 By contrast, the right of exit stipulated in Article 14 of the Withdrawal Agreement is exclusively a right of exit from the host State. While restrictions on the entry of nationals of other Member States are relatively commonplace, restrictions on exit, whether of one's own nationals or other nationals, are less common. There are, however, a handful of cases decided by the Court of Justice concerning exit bans imposed by a Member States on its nationals and ostensibly motivated by public policy considerations. 32 The exclusion of exit controls from the host State in the Withdrawal Agreement would imply that if, for instance, UK nationals residing in an EU27 Member State were on a short visit to the UK, UK authorities could restrict the return of those UK nationals to their EU27 Member State of residence without any possibility to invoke provisions in the Withdrawal Agreement to challenge the restriction on exit.
That said, such exit restrictions would normally be motivated on public policy grounds and, under Article 20 of the Withdrawal Agreement, the use of a public policy or public security derogation in relation to conduct occurring after the end of the transition period will be determined on the basis of national law only and will no longer be subject to control under Chapter VI of the Citizens' Rights Directive. Whether the right of exit from the Home State were to fall within the scope of the Agreement or not would therefore not make much difference since the Withdrawal Agreement would leave the UK or the EU27 Member State concerned in any event largely unconstrained with respect to restrictions to cross-border movements between the UK and the EU based on public policy or public security.
Right of entry and residence. As is the case with the right of exit, Articles 13 and 14 of the Withdrawal Agreement only provide for a right of residence in the host State, not in the home State.
The non-recognition in the Withdrawal Agreement of a right of entry and residence in the home State is less of an issue for the State's own nationals, given the international law obligation on a State to accept its own nationals, 33 than it is for their family members and carers. As is well-known, the Court of Justice has recognised in the Surinder Singh case that family members of EU nationals returning to their home State after having exercised free movement rights in another Member State had a right of entry and residence in the home State at least equivalent to the right of entry and residence they would have in a host State. 34 While the Withdrawal Agreement is quite fastidious in attempting to identify and list in Article 13 every possible configuration of residence rights for Union citizens and their family members in the host State arising out of either Treaty provisions or those of the Citizens' Rights Directive and while it goes as far as codifying in the definition of family members in Article 9(a)(ii) the case law of the Court of Justice extending derived rights of residence to carers on which a Union citizen is dependent, 35 there is no mention at all of any rights of residence of family members on returning to the home State.

Exercise of an economic activity
The Withdrawal Agreement contains two sets of provisions of direct relevance to the exercise of an economic activity by citizens: the first set concerns the package of rights attaching to the status of worker or self-employed person under Articles 45 and 49 TFEU and the second concerns the question of mutual recognition of professional qualifications.
Right to engage in an economic activity and ancillary rights. Articles 24 and 25 of the Withdrawal Agreements transpose within the scope of the Agreement the rights for employed and selfemployed persons contained in Articles 45 and 49 TFEU 36 as well as, with regards to workers, Regulation 492/2011. 37 Article 22 of the Agreement also confirms the right of family members to take up an employed or self-employed activity in accordance with Article 23 of the Citizens' Rights Directive. Here too, however, these rights are expressly stated in the Agreement to be rights exercisable in the host State 38 with no consideration being given to the exercise of economic activities in the home State.
This could potentially open the door to discriminatory practices against EU27 or UK nationals, or their families, by their own Member State, either on returning to their home State or in the situation where they exercise economic activities concurrently in one or more EU27 Member States and in the UK. One could imagine, for instance, a situation similar to that arising in Scholz 39 or Schö ning-Kougebetopoulou, 40 where the home State recognises previous periods of employment in its own public service but not in that of the host State to determine access to employment in the public service or remuneration. Similarly, EU27 nationals residing in the UK, or vice-versa, but who also have economic activities in their home State could conceivably be subject to discriminatory taxation in that State. Yet, if the rights granted by Article 45 and 49 TFEU are, in the context of Articles 24 and 25 of the Withdrawal Agreement, only exercisable against the host State, those individuals may not be able to invoke the benefit of the substantial body of case law developed by the Court of Justice in Schumacker 41 and subsequent cases regarding discriminatory direct taxation of individuals who derive some or all of their income in a Member State other than that in which they reside. 42 Professional qualifications. The Withdrawal Agreement adopts a rather minimalist approach to the mutual recognition of professional qualifications. It only provides for the continued recognition by the host State of qualifications that have already been recognised, 43 or are in the process of being recognised, 44 by the end of the transition period. Even where a qualification was obtained before the end of the transition period, the Agreement does not require recognition by the host State, let alone the home State, if the procedure for recognition was not at least started prior to the end of that period.
In theory, individuals could thus find themselves trapped, from a professional point of view, in the host State, unable to return, or extend their activities, to their home State, to practice their profession without having to re-qualify there for the exercise of that profession.

Social security and social assistance
Social security and social assistance offer a contrasting picture. In relation to social assistance, we find once again the same pattern in the Withdrawal Agreement of entitlement in the host State but not in the home State that we identified above in relation to other rights. In relation to social security, however, the Withdrawal Agreement moves away from an approach in terms of rights to be exercised against the host State and opts instead for the application of the current social security coordination regime more or less in its entirety including in its implications with respect to obligations of the home State towards its own citizens and their families. and limitations 47 to equal treatment that currently exist under Article 24 of the Directive would also apply in the context of the Agreement. Under current EU law, citizens returning to their home State and their families would also in principle be entitled to equal access to social assistance and able to challenge measures and practices of their home State restricting access to social assistance in a manner which discriminates against those who exercised free movement rights in another Member State. 48 Again, this is not an issue that is addressed in the Withdrawal Agreement, which concerns itself exclusively with equal treatment in the host State and not the home State.
Social security. Title III on social security coordination in the Withdrawal Agreement stands apart from the rest of the provisions on citizens' rights in at least two respects: firstly, unlike Title II, it does not express the rights that it confers in terms of rights to be exercised against the host State. Instead, it provides more or less for the wholesale application of EU rules relating to social security coordination 49 to persons covered by the Title. 50 Secondly, the scope of Title III is wider than the rest of the provisions on citizens' rights. Like the general rules in Article 10 of the Agreement, the rules on social security coordination are premised on the individual, being and continuing to be, in a cross-border situation between the UK and the EU27. However, whereas the cross-border element is defined in Article 10 on the basis of two variables (nationality and residence or work), the provisions on social security add a further variable, which is that of the applicable social security legislation.
Thus, EU27 nationals 51 fall within the scope of the Title on social security coordination if they are subject to UK legislation at the end of the transition period. 52 Alternatively, if they are subject to the social security legislation of an EU27 Member State at that point, they also fall within the scope of the Title if they either reside 53 or work 54 in an employed or self-employed capacity in the UK. The same applies mutatis mutandis to UK nationals who are, at the end of the transition period, subject to the legislation of an EU27 Member State or subject to UK legislation while residing or working in an EU27 Member State. 55 As regards continuity, the requirement in Article 30(2) of the Agreement is that a cross-border element as defined above is maintained without interruption, although it does not have to be the same one as long as the individual continues to fall within one of the various situations envisaged in Article 30(1). For instance, EU27 nationals subject to UK legislation would still remain within 47. In particular, economically inactive individuals whose right of residence derives from Article 7(1) the scope of the Title if they became subject to the legislation of an EU27 Member State instead of UK legislation but were nonetheless resident in the UK. 56 Even where someone does not fall, or no longer falls, any of the situations envisaged in Article 30(1), the Title will apply in relation to social security entitlements and obligations resulting from earlier periods of insurance, work or residence both before and after the end of the transition period. 57 Since the rules are not expressed in terms of rights against the host State as such, they will bind the home State just as much as the host State or, for that matter, any other EU27 Member State. Thus, for instance, the UK would not be able to require social security contributions from a UK national working in a self-employed capacity in the UK and in an employed capacity in an EU27 Member State since Article 13(3) of Regulation 883/2004 stipulates that an individual in this situation is subject only to the legislation of the State of employment. Similarly, an EU27 national retiring in his home Member State after having worked in the UK would remain, after the end of the transition period, entitled in her home Member State to medical treatment under that State's social security scheme, in accordance with Article 24 of Regulation 883/2004. Remarkably, Article 32(d) of the Agreement even provides for the payment of family benefits to family members residing in another Member State even in the case of UK nationals residing in the UK and subject to UK legislation, who therefore do not in principle fall within the scope of the Title as defined in Article 31. This is notable as this has been a politically sensitive issue in the UK. 58 The approach in Title III on social security coordination is thus very different from that in Title I and II. This is welcome but we should probably see this as a reflection of the technical complexity and difficulty in untangling rules in this field. The rules are designed to coordinate national social security systems. Their focus is therefore on interactions between those systems. An approach in the Withdrawal Agreement in terms of rights against the host State alone would not fit well in this regime and would add another layer of complexity in a coordination framework which is already anything but simple. The desire to avoid increasing the complexity of the system to the point of unmanageability may also explain why the UK accepted that social security coordination in the context of UK-EU27 relations should remain aligned, subject to limited exceptions, 59 with the internal EU social security coordination system including any amendments occurring after the end of the transition period. 60 56. This could happen, for instance, if an EU27 national residing in the UK quit their job in the UK to take up employment in an EU27 Member State while remaining resident in the UK.

Judicial protection of acquired rights after Brexit
If political institutions have largely failed to explicitly address the question of citizens' rights against their own Member State after Brexit in the Withdrawal Agreement, could courts take over and the legal process provide the protection that the political process has been unable to offer? After all, the protection of citizens' rights against their own Member State in Union law was also largely extrapolated by the Court of Justice out of norms that did not explicitly provide for such protection. This could take the form of a constructive interpretation of the Withdrawal Agreement, Even in the absence of a withdrawal agreement, EU27 nationals residing or having resided in the UK could seek protection of their acquired rights through reliance on their status as Union citizens by extrapolation of the Lounes 61 case law. The situation of UK nationals residing or having resided in an EU27 Member State may be somewhat less favourable. However, the possibility of invoking the Kuric´6 2 case law of the European Court of Human Rights may provide some degree of protection of acquired rights.

A. Constructive interpretation of a withdrawal agreement
The absence of an explicit recognition of the rights of Union citizens against their own Member States in a withdrawal agreement does not mean that such rights could not be read into it. After all, those rights were first developed by the Court of Justice also in the context of an absence of explicit recognition of those rights in the EU Treaties.
One could argue, in effect following the Surinder Singh logic, that recognising (implicit) rights to Union citizens who return to their home Member State after having residing in the United Kingdom before and after the end of the transition period is necessary to avoid them being deterred from exercising (explicit) rights in the Withdrawal Agreement to remain in the UK after the end of the transition period. For instance, EU27 nationals residing in the UK would currently have the right, on return to their home Member State, to be accompanied by their family members regardless of the nationality or nationalities of the latter. If they risked losing that right after Brexit, there would clearly be a disincentive for EU27 nationals to stay in the UK. Similarly, there would be a disincentive for an EU27 national to go to the UK to acquire a professional qualification and/or to stay in the UK after Brexit if, in the case of return to their home Member State after Brexit, that professional qualification was not recognised in their home Member State. The same principles apply mutatis mutandis with respect to UK nationals returning from an EU27 Member State to the UK after Brexit with respect to the right of residence of their family members on return to in the UK or recognition in the UK of a qualification acquired in an EU27 Member State.
Thus, even if the Withdrawal Agreement does not explicitly recognise citizens' right against their own Member State on return to that Member State, such rights could be considered to be an implicit corollary of the rights contained in the Agreement.
That the Court of Justice would take such a line cannot, however, be taken for granted. When it comes to interpreting international agreements to which the Union is a party, the Court has been particularly insistent that the context of the agreement, notably its objectives, its spirit and its 'general scheme' have to be taken into account. 63 The fact that a particular approach to interpretation prevails in the context of the provisions on citizenship in the Treaty on the Functioning of the European Union does not ipso facto signify that a similar approach should be adopted when interpreting a different kind of instrument such as a withdrawal agreement. 64 The fact that citizenship of the Union has been affirmed by the Court to be 'destined to be the fundamental status of nationals of the Member States' 65 may call for a particularly wide interpretation which may not necessarily be warranted when that status is, in effect, being withdrawn (in the case of UK nationals) or rendered inapplicable to the situation at stake (in the case of EU27 nationals). On the other hand, it could be argued that the purpose of the citizens' rights provisions in the Withdrawal Agreement is precisely to neutralise as far as possible the effects of Brexit on the individuals concerned. From this perspective, an interpretation of the scope and extent of those provisions that would seek to mirror the scope and extent of citizens' rights prior to Brexit as closely as possible, including towards their own Member State, may be particularly appropriate. 66 Even if the Court of Justice were to transpose the Surinder Singh logic to the Withdrawal Agreement, there is no guarantee that UK courts would necessarily do the same. There is no obligation in the Withdrawal Agreement on UK Courts to align themselves to the case law of the Court of Justice handed down after the end of the transition period. 67 Nor is there a duty on them to refer questions of interpretation of the Agreement to the Court of Justice beyond the transition period. 68

B. UK nationals: The relevance of Article 8 ECHR and the Kurić case law
Setting aside the provisions that could be contained in a withdrawal agreement, one of the effects of the UK leaving the European Union will be, of course, that Union law will cease to apply to the UK qua Union law. While much Union law would remain, at least provisionally, applicable in substance as 'retained EU Law' under the European Union (Withdrawal) Act 2018, its character as Union law would be lost through that patriation and it would, in effect, become UK domestic law. Deprived of any primacy over domestic law, retained EU law could not protect UK nationals from future interference by the UK legislator with their erstwhile Union citizens' rights.
With respect to residence rights, the European Convention on Human Rights, in particular Article 8 on private and family life, may offer some solace. While the Convention does not per se guarantee a right of entry and residence in a State of which one is not a national and does recognise a general right of States to control the entry, residence and expulsion of aliens, it nonetheless places some limits to the circumstances in which a State may expel a non-national who has been lawfully resident on its territory. In particular, it is clear in the light of the if it were to expel at will a non-UK national who is clearly settled in the UK and has therefore established a private and family life there. 71 The Kuric´case law is of significance mainly to EU27 nationals settled in the UK, or UK nationals settled in an EU27 Member State. Its relevance for UK nationals in the UK is not immediately obvious, as individuals already enjoy in international law a right of residence in the State of which they are nationals 72 and therefore do not need to rely on Kuric´to establish that right of residence. It would, however, be relevant for family members who do not have the nationality of the home State. In other words, it could be relevant for Zambrano carers of UK nationals and for non-British family members of returning UK nationals in a Surinder Singh situation. As long as the family is already settled in the UK before the end of the transition period, the non-British members of the family could rely on that case law to oppose an attempt by UK authorities to call into question their right of residence in the UK after the end of the transition period.
Vidmar suggests that Kuric´concerns primarily non-nationals who have established permanent residency in the host State. 73 It is debatable whether the scope of Kuric´is limited to permanent residents. The application of Article 8, however, would presuppose that the individual has already established a private and/or family life in the UK. If the family has only recently moved to or formed in the UK, the application of Kuric´seems doubtful as there would not be a sufficiently well-established private and family life in the UK to be protected under Article 8.
Beyond Article 8 and the Kuric´case law, it is not immediately clear that there is much in the ECHR to address the issues specifically raised by Brexit with respect to the preservation of Union citizens' rights enjoyed by UK nationals and their families vis-à-vis the UK. To take the examples of taxation and mutual recognition mentioned above, neither the ECHR -or indeed any other norm -seems apt to protect a British national returning from an EU27 Member State to the UK after Brexit from a refusal by UK authorities to recognise a professional qualification acquired in that EU27 Member State or to protect a British national residing in an EU27 Member State after Brexit from taxation rules that discriminate against non-UK residents even where such refusal of recognition or taxation rules would constitute a breach of Union citizens' rights when exercised in the context of intra-EU cross-border movements.
C. EU27 nationals: The persistence of Union citizen status and the Lounes case law EU27 nationals are in a different situation since they will remain Union citizens after Brexit and will therefore be in a position to invoke that status towards their own Member States. The difficulty for them, however, lies in the fact that the exercise of most Union citizenship rights is premised on having exercised right to free movement in the Union. If the UK is no longer a Member State of the Union, cross-border movement with the UK will no longer constitute the exercise of EU free movement rights. It seems clear, for instance, that an EU27 national moving to the UK after Brexit could not, on a future return to his or her Member State of origin, invoke the benefit of the Surinder Singh case law. Nor could such an individual challenge any discriminatory treatment by the home Member State on account of the individual residing or working in the UK rather than the Member State of origin or, more generally, the Union. Union law on free movement protects the free movement of persons between Member States within the European Union but does not protect the free movement of persons between the EU and third countries and the movement of a person from the EU27 to the UK would be a movement from the EU to a third country and therefore unprotected by EU free movement law.
The same, however, may not be true of EU27 nationals who are residing or working in the UK at the time of Brexit or at the end of the transition period and continue to reside or work in the UK after that. Those individuals have exercised rights to free movement in the Union. The fact that the UK leaves the Union so that the place where they have exercised their free movement rights is no longer within the European Union does not change the fact that EU free movement rights were originally exercised. From this perspective, we are here in a Surinder Singh-type of situation: The right to return to the home State from the UK with one's family is protected because it is the corollary of the original outbound right to free movement to the UK and it is that original outbound right to free movement which is, in effect, indirectly protected. If that is so, there does not seem to be any reason why the departure of the UK from the European Union should lead us to refrain from protecting that right as long as the outbound right was exercised at a time when the UK was still in the Union.
We could draw an analogy here with the decision of the Court of Justice in Lounes. 74 In that case, the Court found that the right of residence in the UK of a Spanish national who had also obtained British nationality by naturalisation after moving there had an unconditional right of residence by virtue of her British nationality rather than by virtue of Union law. Nonetheless, by moving to the UK as a Spanish national, she had exercised rights to free movement and came, as a result of that, within the ambit of Union free movement law. Those free movement rights include the right to lead a normal family life together with their family members in the host State, which could not be lost as a result of the later acquisition of British nationality by the individual concerned. As a result, Mrs Lounes's third country national spouse enjoyed a right of residence in the UK for the purpose of allowing Mrs Lounes to enjoy her right to free movement and to the enjoyment of a normal family life in the host State by virtue of Article 21(1) TFEU.
In the Lounes case, therefore, the change in the basis of her right of residence in the UK from EU Law to national and international law did not alter the fact that the original right to free movement was still protected even after the change of immigration status of Mrs Lounes. The fact that Mrs Lounes had exercised rights to free movement to enter and reside in the UK as a Spanish national 74. Case C-165/16 Lounes.
was crucial, otherwise we would have been in an identical situation to that in McCarthy, 75 in which the dual nationality of Mrs McCarthy as Irish and British citizen was not in itself enough to bring her within the ambit of Article 21 TFEU to generate a right of residence in the UK for her third country national spouse. What is being protected in Lounes therefore is the original exercise of the free movement right.
In the same way as Mrs Lounes, following the acquisition of British nationality, was no longer exercising her EU law rights to free movement when residing in the UK, EU27 nationals would not, after the end of the transition period, be exercising their EU Law rights to free movement when residing in the UK. Yet, in the same way as Mrs Lounes original exercise of the right to free movement had to be protected by continuing that protection even after she ceased to exercise it by acquiring British nationality, the original right to free movement to the UK by EU27 nationals would need to be protected by continuing that protection even after they cease to exercise it following the departure of the UK from the European Union.
Admittedly, the Court in Lounes noted that the acquisition of British nationality was in addition to Spanish nationality, leaving open the possibility that the outcome might not have been the same had Mrs Lounes lost her Spanish nationality on acquisition of her British one. Mrs Lounes was still a Spanish national residing in another Member State even though she was at the same time a British national residing in her home State. The analogy between Lounes and the post-Brexit situation breaks down here: one can be a dual Spanish and UK national but the UK cannot be in and out of the Union at the same time. That said, for the outcome in Lounes to depend on whether Mrs Lounes retained her Spanish nationality or not would seem difficult to justify. First, it would introduce arbitrary discriminations between EU nationals depending on whether their country of origin allows for dual nationality or not. Secondly, the rationale behind Lounes is 'to ensure that the Union citizen can exercise his freedom of movement effectively.' 76 Thus, whether the individual is still a national of the Member State of origin should not be material for the application of the Lounes case law. Similarly, the fact that the UK is no longer part of the EU should not be determinant as long as the rights were originally exercised while the UK was in the Union.
If this analysis is correct, it would mean, in particular that an EU27 national who has been continually residing in the UK before and after Brexit and then returns to their member State of origin should be in a position to invoke the benefit of the Surinder Singh case law and invoke a Union law-based right of residence for their spouse and children. 77 Beyond migration rights and taking the examples of taxation and mutual recognition discussed in the preceding section concerning UK nationals, EU27 nationals having continually resided in the UK before and after Brexit could arguably, when later returning to their home Member State, insist on recognition by their home Member State of a qualification acquired in the UK. Similarly, EU27 nationals residing in the UK could object to taxation by their home Member State that would discriminate against them on the basis of their residence in the UK. Compared to UK nationals, EU27 nationals would therefore potentially be in a somewhat stronger position vis-à-vis their own Member State: transposition of the Lounes case law would enable them to future-proof the consequences after Brexit of 75. Case C-434/09 McCarthy. 76. Case C-165/16 Lounes, para. 48. 77. This would remain true as long as the move to the UK takes place before the end of the transition period or, in the case where no withdrawal agreement is concluded, the date of withdrawal from the Union: as long as the right to move to the UK exists, the failure to recognise a Surinder Singh right to return in the future would undermine the outbound right to move by creating a disincentive to exercise it.
having exercised free movement rights prior to Brexit, a result that cannot be achieved, or at least not to the same extent, under the Kuric´case law, whose main focus is to protect a right of residence in the host state to individuals already settled there. All in all, there is therefore a strong case for considering that many of the rights enjoyed by Union citizens against their own Member State should continue to be at least partly protected after Brexit for individuals who have exercised free movement rights between the UK and the EU27 Member States at the time of Brexit or the end of the transition period and continue to exercise such rights afterwards. If the Withdrawal Agreement is ratified, this could be done by a constructive interpretation of the Agreement. In the absence of an Agreement, reliance on the Lounes case and the Surinder Singh case law on the EU27 side would allow a very similar outcome to be reached. On the UK side, however, the protection provided by Article 8 of the ECHR would be limited to the recognition of some residence rights and the persistence of rights or advantages for UK nationals deriving from their former status as Union citizens more dependent on benign treatment under UK domestic law.

The structure of the Article 50 process and the role of the institutions in it
Article 50(2) contemplates the conclusion of an agreement between the Union and the leaving Member State setting out the arrangements for the withdrawal of that Member State. The expected coverage of the withdrawal agreement is thus left rather vague in Article 50(2). Those 'arrangements' could however prima facie reasonably include the consequences of withdrawal on the rights enjoyed by Union citizens -including those of the soon-to-leave Member State -and the Member State of which they are a national. Even though some aspects of the withdrawal procedure, notably the participation of the European Parliament, could be seen as opening a window of opportunity for consideration of this issue, the centrality of negotiations to the withdrawal process constitutes a strong hindrance.

A. EU withdrawal as a constitutionally-structured negotiation process
The introduction of an explicit withdrawal clause in the Treaties following the Lisbon Treaty had two consequences which, while linked, nonetheless give rise to potential misalignments or tensions: first, it placed the withdrawal process resolutely in an EU constitutional framework rather than that of general international law. 78 Second, the choice of procedure for withdrawal made in Article 50 TEU puts negotiation between the EU institutions and the remaining Member States on the one hand and the leaving Member State on the other at the heart of that process.
Those two consequences are linked, in that the very choice of a withdrawal procedure centred on negotiations with the departing State is not typical of international law practice which, as Hillion notes, is often limited to requiring a mere notification of the intention to withdraw. 79 In that sense, the presence of a negotiated procedure could be seen as signalling a move away from standard international law. Yet, the very format of a negotiation between the departing Member State and the international organisation it is leaving adopts methods and tools which are those of international inter-governmental processes rather than those which are typical of domestic constitutional settlements. Indeed, Article 50(1) TFEU provides that the withdrawal agreement is negotiated in accordance with Article 218(3) TFEU, viz. following the format for negotiations of international agreements between the Union and third countries or international organisations. Thus, the withdrawal process consists at heart of carrying out an international negotiation for the adoption of an international treaty albeit within an EU constitutional framework rather than the standard international law framework.
The centrality of international negotiations gives the withdrawal process a strong adversarial character in which each side fights to protect the interests of its own camp and extract concessions from the other side. While considerations of the rights of individuals can figure prominently in such negotiations and this has indeed been the case in the Brexit process, the focus is nonetheless on one side protecting its citizens against unfavourable treatment by the other side. It does not lend itself to thorough examination of the problem that concerns us here, namely that of the unfavourable treatment by a Member State, or former Member State, of its own citizens.
Eeckhout and Frantziou have argued that the fact that article 50 TEU is situated in a 'quintessentially constitutional place' militates in favour of what they call a 'constitutionalist reading' of it. 80 Among other consequences, such a reading, they contend, would demand that the withdrawal agreement should guarantee the protection of fundamental rights relating to private and family life as a crucial condition of its constitutionality 81 and possibly even contain measures to guard against regression in the level of protection of other acquired rights. 82 If Eeckhout and Frantziou are right, this would potentially open the door for an argument that the constitutionalist reading they argue for could require the withdrawal Treaty to contain measures to protect Union citizens against loss of rights not just at the behest of the other side in the negotiations but also at the behest of the Member State of which they are a national. Protection of citizens' rights would be the result not just of one side bargaining hard to protect its people from unwelcome treatment by the other side but the reflection of a constitutional imperative. As such, there is no reason a priori to exclude the possibility that those rights would include rights to be exercised against one's own Member State.
As Eeckhout and Frantziou themselves recognize, however, it would be difficult to conceive of a general legal obligation on the EU or the UK to protect all rights acquired as Union citizens where there can be no legitimate expectation that a Member State will always remain in the Union. Even in relation to the protection of the right to private and family life or, for that matter, other fundamental rights, it is difficult to see how there could be an obligation to ensure such fundamental rights protection in the withdrawal agreement when there is no obligation to conclude the agreement in the first place. If a 'constitutionalist reading' of Article 50 TEU required rights to private and family life of Union citizens to be protected and that protection must be included in the withdrawal agreement to be ensured, it would seem to follow that there should be an obligation to conclude such an agreement as a condition of withdrawal. Yet we know that such obligation does not exist, since Article 50(3) TEU contemplates a withdrawal of the Member State even in the absence of such an agreement. 80 This is not to say that no obligations to protect acquired rights exist. As was discussed above, there is a strong case for considering some obligations of this kind, derived from either international human rights law, or even EU law itself as far as the EU27 are concerned, do exist. Rather, the point is that those obligations are unrelated and owe nothing to the EU constitutional context of the withdrawal process and would arise in identical terms in a purely international law-driven withdrawal process. The existence of such obligations does not therefore detract from the point made here about the withdrawal process not being an environment conducive to bringing to the fore the issue of the impact of leaving the Union on the relationship between individuals and the State of which they are a national, whether on the side of the leaving Member State or the remaining ones.
B. The role of institutional actors in the withdrawal process As noted above, the negotiation of the withdrawal agreement is the central piece of the withdrawal process and that negotiation is based on the format specified in Article 218 TFEU with one notable, difference, which is the role of the European Council. The wording of Article 50 TFEU suggests a role for the European Council limited to the issuance of initial negotiating guidelines and that of agreeing, if necessary an extension to the negotiation period in agreement with the withdrawing Member State. In the context of the Brexit negotiations, however, the European Council has had a far more involved role than the wording of Article 50 might suggest. Even before the Article 50 process was formally started by the notification by the UK of its intention to withdraw from the Union, the Heads of State or Government of the 27 together with the Presidents of the European Council and of the Commission had already indicated at an informal meeting in December 2016 that the European Council would 'remain permanently seized of the matter' and would update the guidelines 'in the course of the negotiations as necessary.' 83 This monitoring of the European Council throughout the process enabled to exercise considerable control over not just the substantive contents of the negotiations but also its sequencing. 84 Apart from the involvement of the European Council, the pattern followed for the Brexit negotiations was in line with the usual Article 218 TFEU framework with the Commission negotiating on the basis of a Council mandate and Council directives and the Council concluding the agreement.
The European Council and the Council would be expected to approach the withdrawal from the perspective of its impact on national interests or on the interest of the EU as a whole albeit seen through the prism of national interests. From such a standpoint, for the Member States to decide to impose obligations on themselves collectively to protect their own citizens against actions that the Member States themselves might take following Brexit seems to run against the grain. Clearly, here too the focus will be on imposing obligations on the withdrawing Member State to protect the citizens of the remaining Member States and conversely accepting obligations towards the citizens of the withdrawing Member State in return. The Commission might have a slightly different perspective and may be somewhat more inclined a priori to consider the interests of Union citizens 83 in a more holistic manner, including in terms of its implications in the relationship between Union citizens and their own Member State. Nonetheless, the dynamics of the negotiation, not to mention the negotiating mandate and the guidelines and directives from the European Council and Council are also going to lead to a focus, as far as citizens' rights are concerned, on mutual obligations towards the citizens of the party on the other side of the negotiating table rather than obligations towards citizens on one's own side.
If there is one institution involved in the withdrawal process whose perspective is likely to be more centred on the rights of Union citizens qua citizens, it would have to be the European Parliament. According to the letter of Article 50, the involvement of the European Parliament is limited to giving (or withholding) its consent to the withdrawal agreement. One could perhaps argue that Article 218(10) TFEU, which gives the right to Parliament in the context of negotiations of international agreements to be 'immediately and fully informed at all stages of the procedure' should apply by analogy. 85 In any event, the power to give or withhold consent gives the European Parliament leverage to make its voice heard during the negotiation process. This was acknowledged in the statement by the Heads of State or Government of the 27 at their December 2016 informal meeting, which anticipated that the Commission, as Union negotiator, would 'keep the European Parliament closely and regularly informed throughout the negotiation' and that the President of the European Parliament would be invited to be heard at the beginning of European Council Brexit meetings. 86 De facto, the European Parliament, in particular through its Brexit Steering Group, has been regularly consulted by the EU Chief Negotiator and has also issued a number of resolutions on the Brexit negotiations. 87 There is no doubt that citizens' rights figure prominently in the positions expressed by the European Parliament. What is particularly interesting in that context is that the Parliament adopts a rather less partisan approach to the issue and shows concern about the rights of all Union citizens including UK nationals rather than just seeing its role as defending the rights of EU27 citizens. It is in this respect quite telling that the Parliament has repeatedly called for UK nationals falling within the scope of the Withdrawal Agreement to be recognised rights of residence and free movement after Brexit not just in the specific Member State in which they continue to reside after the end of the post-Brexit transition period but throughout the whole of the territory of the Union. 88 There is, in this respect, a marked difference between the positions of the EU27 European Council and Council and of the Commission on the one hand and that of the European Parliament on the other: Whereas the former see themselves as essentially defending the interests of their side in the negotiations, the European Parliament adopts a more holistic approach to the interests of all citizens across the UK/EU27 divide, even where it means imposing additional obligations on the EU27 side rather than the UK side.
Yet, there are no references to the impact Brexit might have on the rights enjoyed by citizens vis-à-vis their own Member State in the various resolutions adopted by the European Parliament and in the statements issued by its Brexit Steering Group. Admittedly, Parliament has on multiple occasions insisted that Brexit should not result in a deterioration of the position of Union citizens and that the Withdrawal Agreement should incorporate the full set of rights currently enjoyed by Union citizens. 89 While Parliament draws the consequences of this in terms of commitments of the UK and of the EU27 towards each other's citizens, it does not do so in terms of commitments of both sides towards their own citizens.

Conclusion
The importance of safeguarding citizens' rights was recognised by all parties involved in the Brexit process. Despite this, the Withdrawal Agreement leaves a number of gaps in the protection of those rights. A particular area of concern that this article has sought to address is that of the rights of citizens against the State of which they are nationals. As discussed above, the fact that continuous residence in the host State is required to remain under the protective umbrella of the Agreement means that individuals risk losing that protection when they return home, at the very time they are likely to need it most with respect to their own State. It also means that those who return early, before the end of the transition period, will find themselves entirely unprotected by the Agreement even with respect to the consequences of a situation that might have arisen during the exercise of their free movement rights in another Member State. 90 Moreover, the way most citizens' rights other than those deriving from social security coordination are defined in the Agreement as rights to be exercised 'in the host State' would seem to preclude their invocation against the home State, leaving the door open to discriminatory practices in such fields as, for instance, taxation or social assistance against returnees or those whose economic activities span across the UK/EU27 divide, to non-recognition by the home State of qualifications obtained in another Member State or to restrictions or denials of residence rights or access to employment to family members.
That said, the failure of the Withdrawal Agreement to actively and explicitly engage with the rights of citizens against their own State is not necessarily fatal to the protection of those rights. It would not be the first time in the history of European integration that courts step in to fill a void left by political processes. First and foremost, on the assumption that the current Withdrawal Agreement or a variation of it, is eventually ratified by both sides, Courts could take their cues from the Surinder Singh case itself to adopt a constructive interpretation of the provisions of the Agreement so as to ensure the full effectiveness of the rights that it contains and avoid any disincentive to the exercise of these rights. This would imply ensuring that individuals are no worse off returning to their home State after exercise of the right in the Agreement to stay in the host State than they would be not exercising that right and returning home prior to Brexit. It would also imply that an individual should not be discriminated if they exercise economic activities that span the EU27/UK divide after the end of the transition period. Thus, courts should be able to read into the agreement much of the protection currently provided by EU law in the relationship between EU nationals and their own Member State, thereby avoiding the negative consequences outlined in the previous paragraph. Even if the courts were to decline adopting such an interpretation of the Withdrawal Agreement, or if no withdrawal agreement were ratified, we saw that there would still be room to offer at least some protection of acquired rights in the UK based on Article 8 ECHR for families in Zambrano or Surinder Singh types of situations as long as the family had lived long enough in the UK to be regarded as having a private and/or family life there. On the side of EU27 Member States, reliance on the Surinder Singh case law should still remain possible after Brexit even in the absence of a Withdrawal Agreement, especially in the light of the willingness of the Court of Justice in Lounes to continue protecting the exercise of a right to free movement even where what constituted the exercise that right can no longer be regarded as such an exercise due to a change in legal status. 91 Even if there is therefore space for courts to compensate for or minimise the non-engagement of the Withdrawal Agreement with the rights of citizens towards their own Member States, this lack of engagement remains nonetheless problematic. Eeckhout and Frantziou rightly urge us to adopt a constitutional reading of Article 50 TEU. 92 Such a reading entails the protection of citizens' rights. As they point out, a 'meaningful constitutional interpretation of Article 50 indeed requires in-depth consideration of respect for individual rights as one of the most settled features of the EU constitutional order to date'. 93 This includes avoiding a purely transactional approach to the withdrawal process and taking citizens' rights seriously. Not treating citizens as 'bargaining chips' is one aspect of that duty to take citizens' rights seriously. 94 Going beyond extracting concessions from the other side to protect one's citizens and also asking oneself what one needs to do towards one's own citizens to protect those rights threatened by the withdrawal process is also part of it. To that extent, the lack of engagement in the Brexit negotiations with the issue of the rights of Union citizens vis-à-vis their own Member State is at least a partial failure to adopt a constitutional reading of Article 50.
As discussed in this paper, the adversarial structure of a process focused on negotiations between opposing parties has certainly contributed to this. This has to some extent been mitigated by the involvement of the European Parliament but these is something of a missed opportunity here for Parliament to assert itself more robustly as an impartial protector of the rights of all Union citizens by bringing in this additional dimension of the citizens' rights debate in the Brexit negotiations.