Abstract: What is the point of expelling illegally residing EU citizens when they would be able to immediately return in an open-border Europe? This Insight critically discusses the opinion by Advocate-General Rantos in the FS v Staatssecretaris van Justitie en Veiligheid case (C‑719/19), where an EU citizen had been expelled from the Netherlands for lacking employment and sufficient resources but returned a month later to claim a renewed right of residence under art. 6 of the Citizenship Directive. It is argued that the Advocate-General’s line of reasoning and solution for the problem should have shown more awareness of existing legal doctrine and the actual functioning of Member State administrations. Specific criticism is directed at his attempt to frame “police resources” as part of the social assistance system, his return to formalism in the adjudication of EU citizens’ rights and his use of (another) “case-by-case” assessment as a solution to the problem. A presumptive threshold of three months absence seems least bad amongst alternatives to assess whether EU citizens comply with an expulsion decision, especially in light of clarity and legal certainty for both EU citizens and national authorities.
Keywords: expulsion – EU citizenship – sufficient resources – return – Directive 2004/38 – case C‑719/19.
What is the point of expelling illegally residing EU citizens when they would be able to immediately return in an open-border Europe? This, in rather rough terms, is what is at stake in the case of FS, a Polish citizen who was expelled from the Netherlands for lack of employment and sufficient resources but who returned a month later. Would he be able to rely again on art. 6 of Directive 2003/38 (“Citizenship Directive”) – granting every EU citizen a right to reside in another Member State for three months without conditions – now that he has left that Member State’s territory and “re-enters” again? Or, rather, would Member States be able to prolong the effects of such an expulsion decision as, otherwise, the tools Member States have under the Citizenship Directive to exclude undesired EU citizens from their territory would be rendered largely ineffective? The case goes to the very heart of the question how to administer the internal contradictions of a federalised free movement regime like that of the European Union, where advanced free movement rights and radically open borders are combined with the authority of Member States to expel undesired foreigners. In his proposed solution to the questions, Advocate General (AG) Rantos seems to be aware of the contradictions inherent to the EU’s free movement regime but demonstrates a remarkable lack of understanding of its functioning and administration in practice and an equal lack of understanding of EU law and the jurisprudence of its own Court.
II. Facts of the case
Mr. FS is a Polish citizen living in the Netherlands. He is not a stranger to the Dutch police, as evidenced by 14 notifications on various counts in the police registry. On 1 June 2018, the Dutch immigration service issued FS with a decision which determined that he did not have a right of residence under EU law and ordered him to leave the Dutch territory within 28 days. This decision was not taken “on grounds of public policy or public security” under art. 27 of the Citizenship Directive, but on grounds of his “poverty”, i.e., he did not fulfil the residence conditions under art. 7 of that Directive. Although FS had worked for a period of 5 months in the past, the Dutch immigration service found FS no longer employed and unable to prove his involuntary unemployment. Additionally, FS was unable to demonstrate he had sufficient resources to support himself, which, according to the immigration service, was also evidenced by the number of arrests on the suspicion of shoplifting and pickpocketing. After the decision was made final on 25 September 2018, FS did what was expected of him and left the Dutch territory within 28 days. He resided with friends across the border in Germany, where on 23 October he was arrested for suspicion of shoplifting. A month later, however, he decided to return to the Netherlands because he received an order to appear before a Dutch court. A day before this hearing he was once more arrested for shoplifting and placed in administrative detention with a view to returning him to Poland.
The appeals FS brought against this administrative return decision would end up before the Dutch Council of State and form the subject of the preliminary questions this court referred to the European Court of Justice. FS argued that his administrative return decision was unlawful: by leaving the Netherlands to Germany he had complied with the expulsion order and when returning he had acquired a renewed right to reside in the Netherlands under art. 6 of the Residence Directive. The State Secretary, while admitting that FS had left the Netherlands, argues that the legal effects cannot be exhausted by simply leaving the territory: an expelled Union citizen in the case of FS should at least settle in another Member State under art. 7 of the Citizenship Directive and demonstrate “genuine residence” in that other Member State for a period of more than three months. Otherwise, so the State Secretary keenly remarks, it would be sufficient for an expelled Union citizen like FS to stay in Germany for one day and return and reside legally in the Netherlands once more.
The preliminary questions come down to the question if a Union citizen who receives an expulsion decision on grounds of not satisfying the conditions of art. 7 Citizenship Directive complies with that decision by demonstrably leaving the territory of the expelling Member State. If the answer is yes, does the Union citizen have a renewed right of residence under art. 6(1) of the Citizenship Directive, barring the expelling Member State from expelling him/her again? If the answer is no, should the Union citizen reside outside the territory of the host Member State for a certain period of time and, if so, for how long?
III. Opinion of AG Rantos
In his opinion, Advocate-General Santos first evaluates art. 15 of the Directive. As the Court had already clarified in Chenchooliah, this provision forms the legal basis for expelling EU citizens on “grounds other than public policy, public security or public health”, in particular when EU citizens no longer satisfy the requirements for granting the right of residence under arts 6 and 7 of the Directive. Here it should be recalled that art. 6 provides for a “short-term” right of residence up to three months without any other condition than that EU citizens holds a valid identity card or passport and art. 7 provides for a “mid-term” right of residence for periods longer than three months under the condition that EU citizen are employed or economically self-sufficient. Discussing these articles together in the context of the case, the AG claims that an expulsion decision on the basis of art. 15 can “solely” be justified in relation to the criterion that the EU citizen should not represent an unreasonable burden on the social system of the host Member State. As will be further discussed below, this interpretation poses special difficulties at a later stage in the opinion since FS did not receive or claim welfare benefits. Initially, Rantos skips this problem by stating that art. 15 applies where a right of residence comes to an end because the person concerned “risks” becoming an unreasonable burden on the social assistance system of the host Member State. With that preliminary remark in mind, Rantos answers the first question by concluding that merely physically leaving the territory is not enough to comply with an expulsion decision issued on the basis of art. 15. Not being able to draw conclusions from the text and context of art. 15, he interprets the objective behind this provision as allowing Member States to protect their public finances by expelling EU citizens who represent an unreasonable burden on the social assistance system. Considering EU citizens to comply with an expulsion decision when crossing the borders would, according to the AG, not only deprive art. 15 of its practical effect but also the substantive residence conditions of art. 7.
Having answered the first question in the negative, the AG moves on to discuss the question if EU citizens must reside outside the host Member State for a certain period of time and, if so, for how long. He rejects the position of the Dutch government that the legal effects of an expulsion decision are only exhausted when EU citizens genuinely and actually leave a Member State for a period of three months. While admitting that this interpretation might contribute to the effectiveness of expulsion decisions and legal certainty of both the EU citizen and the Member State, the AG finds the systematic application of such a fixed minimum period in violation of the Directive. The AG provides two reasons for this. First, since EU law does not attach any temporal effect to expulsion decisions, “endorsing the Netherlands Government’s proposal would […] be a judicial creation, as the Court would be taking the place of the EU legislature”, which would not only violate the primary right to free movement but also the principles of institutional balance and the conferral of powers. Secondly, the imposition of a minimum period of absence would introduce an additional condition on the exercise of the right of entry and be perceived as an entry ban, which is prohibited under art. 15(3). The AG considered the Dutch proposal to allow expelled EU citizens to return for “specific and concrete” reasons such as medical check-ups or one-time purchases as “subjective and potentially arbitrary”, thereby deterring EU citizens from using that possibility. Both arguments will be criticized in the Comment section below.
So, what is the alternative presented by the AG? According to Rantos, national authorities should deal with the temporal effects of expulsion decisions on a “case-by-case” basis. He offers three “kinds of factors” that national authorities should take into account when deciding if an expelled EU citizen is welcome again. The first kind of factors relate to the need of the EU citizen to prove that s/he has genuinely and actually ended his/her period of residence in the host Member State. According to the AG, this would be more “demanding” for a mid-term resident under art. 7 – e.g., by showing deregistration, termination of a lease and utilities agreements, etc. – than a short-term resident under art. 6 who “has not had time to become integrated”. The second kind of factors relate to the reasons that have led to the expulsion in the first place. Here, the AG suggests that national authorities should conduct an individual examination as to “whether the expelled person, although not satisfying the conditions of art. 7 of the Residence Directive, no longer represents an unreasonable burden for the social assistance system, and may therefore rely on a right of residence under art. 6 of that directive.” Consequently, he adopts an (extremely) broad definition of social assistance, namely covering “all the benefits that the Union citizen has not contributed towards and which are funded by the public purse” whereby he includes the possibility that an EU citizen, by engaging in “repeated criminal conduct following his or her expulsion”, could represent an unreasonable burden “on account of the mobilisation […] of a significant amount of police resources”. Finally, the AG suggests to take account of “the genuine intention” of an EU citizen to comply with the expulsion decision as a factor, although he admits that this would border “pure supposition” and run into practical difficulties.
The case presents a difficult question of law. Yet in contrast to what the AG claims, the circumstances of the case are not that “unusual”. At least in the Netherlands it has become general practice to issue expulsion decisions and forcefully expel Union citizens on the basis of their lack of sufficient resources even though they do not claim social assistance. In the context of an open border Europe, a significant number actually return and are subjected to another expulsion decision. Also Union citizens claiming social assistance risk receiving a 28-day limit to leave the Dutch territory: when would they be able to rely on a renewed right of residence under art. 6? The problem raised by the Council of State is therefore a genuine one, but the solution provided by the AG is not satisfying for at least the three reasons that are discussed below.
IV.1, Police resources as social assistance?
AG Rantos starts from the assumption that expulsion decisions on the basis of art. 15 can only be justified in relation to EU citizens who represent an unreasonable burden on the social assistance system and builds his subsequent arguments around it. In the factual and legal context of the case he thereby digs his own grave. The EU citizen in question, in fact, had never relied on social assistance and was removed from the Netherlands for not possessing “sufficient resources”. The Dutch authorities found evidence for this, amongst other types of evidence, in his petty-criminal behaviour. While the AG later – quite confusingly – talks about the “risk” of becoming an unreasonable burden (already something completely different from actually becoming an unreasonable burden), this becomes even more problematic when he suggests that national authorities should examine whether an EU citizen “no longer” represents an unreasonable burden at the moment an EU citizen requests a renewed right of residence under art. 6. The problem, of course, is that FS has never been such a burden in the first place (and even if he had, he would probably no longer receive social assistance after his right to residence was terminated).
The AG solves this problem by redefining the concept of “social assistance” as “all the benefits that the Union citizen has not contributed towards and which are funded by the public purse” which could include “police resources” that are mobilised to correct the behaviour of a Union citizen. This assertion is almost beyond parody. It relies on a gross misreading of para. 61 of Brey (which the AG refers to), where the Court has ruled that the: “concept must be interpreted as covering all assistance introduced by the public authorities, […] that can be claimed by an individual who does not have resources sufficient to meet his own basic needs and the needs of his family and who, by reason of that fact, may become a burden on the public finances of the host Member State during his period of residence which could have consequences for the overall level of assistance which may be granted by that State”.
Brey is not an isolated case and the Court repeated this definition in the later Dano (2014) and Alimanovic (2015) judgments. I am not denying that individuals like “FS” might constitute a burden on the “public purse” by shoplifting behaviour or other petty crimes, but the resources invested to correct such behaviour can never be deemed assistance introduced by public authorities to help individuals meet their basic needs. Perhaps, the argument could be made that his behaviour – by putting a strain on the police budget – indirectly affects the overall level of social assistance that could be provided by the host Member State, but this relationship is too tenuous as many economic factors and political choices are in between. It would also place Union citizens engaged in any kind of activities that are a burden on the “public purse” in some way or another – e.g., students demonstrating for Extinction Rebellion – on the same pedigree. Indeed, qualifying police resources as “social assistance” under the Citizenship Directive seems incorrect, especially in light of the specific history of the term “social assistance system” in secondary legislation.
The Court has to make a decision, if not in this case then in another, whether expulsions can actually take place with respect to EU citizens who do not rely on social assistance. For sake of analysis, there might be roughly two solutions. One is to accept that, indeed, EU citizens who do not rely on social assistance should be assumed to legally reside in the Member State and accumulate residence time under the Directive and qualify for permanent residence after five years. This interpretation seems to be the view of AG Whatelet in Gusa (2017) and finds support in recital 16 of the Directive. The downside would be, at least for Member States, that mere “survival” on a State’s territory for five years, even under the social substance level, would be sufficient to acquire permanent residence status and access to the host welfare system. The alternative solution would be to accept the broader idea that the residence conditions – employment or “sufficient resources” – are meant to safeguard that all inhabitants in a territory live a life in accordance with the minimum of social standards agreed upon in that society. EU citizens making ends meet below that line – often expressed in a social assistance norm – are therefore not deemed to meet the sufficient resources norm and, as a result of this, become eligible for social assistance to top up their income to an acceptable standard. This reasoning pleads in favour of accepting Member States to employ a certain threshold and allowing them to expel EU citizens not meeting that threshold irrespective of whether they actually rely on social assistance. This would be in line with the assertion that EU citizens need to have sufficient resources in order to avoid that they become a burden on the social assistance system. It would also be in line with the (confusing) art. 8(4) Citizenship Directive, which seems to allow Member States to require EU citizens to have at least but not more than the social assistance threshold.
IV.2. How many more “individual examinations”?
The AG rejects the Dutch proposal to introduce a three month period of absence before expelled Union citizens are welcome again, instead proposing to national authorities to assess a renewed right of residence on a “case-by-case” assessment.
Remarkable is his assertion that the Court, by endorsing the Dutch solution, would be creating law and take the seat of the Union legislator and thereby act, essentially, ultra vires. This statement surely constitutes a fresh sound from within the Court, but is incorrect in light of the Court’s case law and lacking understanding of the judicial dialogue taking place in the decentralised system of EU law enforcement. Sure, Member States could harmonise the duration of expulsion decisions in secondary legislation, but in light of the often ambiguous and open-ended rules that are the result of compromise between and within EU institutions, Member States have no other choice than to actively interpret EU norms in their domestic context. It is subsequently up the Court to endorse or reject such interpretations, whereby it occasionally endorses temporal thresholds. In the recent G.M.A. case, for example, the Court accepted six months as “a reasonable period of time” to find a job. Calling this “judicial creation” might be a sign of a rebelling Advocate General, but it is far off the mark when it comes to understanding the role of the Court in accepting and rejecting national authorities’ interpretations of Union law. Although, indeed, the Court’s has a tendency to require national authorities to test “individual circumstances”, the Court is not entirely insensitive to the needs of those authorities to work with predictable thresholds in order to ease bureaucratic workload and create legal certainty for both civil servants and Union citizens.
The AG’s alternative is the traditional response of not knowing how to strike a delicate balance between free movement rights with Member States’ prerogative to protect their social assistance systems, namely by proposing another “case-by-case” examination. Of course, obliging Member State authorities to take account of “individual circumstances” before rigidly imposing restrictive measures on EU citizens has helped to extend social rights for “non-traditional” categories of EU citizens. However, introducing individual examinations shifts the burden of administration to Member State authorities and the burden of uncertainty to individual EU citizens. Can one seriously expect from Member State authorities to determine the “genuine and actual nature of the end of the period of residence” and from the Union citizen to deliver all the relevant evidence in that respect? Do we really want to leave it up to Member State authorities to determine “the risk” of an EU citizen becoming an unreasonable burden on the social assistance system or even the original “intention” of that citizen to comply with the expulsion decision? It seems to me that such individual examinations would lead to precisely those concerns the AG expresses elsewhere in his opinion, namely that they are essentially “subjective and potentially arbitrary”, that their very existence might deter Union citizens from applying for a renewed right of residence and that they might entrench the temporal effects of an expulsion decision on an unlimited basis.
IV.3. A strange return to formalism
This brings us to the final point of critique. The AG seems to be keenly aware of the awkward nature of the Union’s free movement regime, noting that Member States have very limited means of verifying that the expulsion of an illegally residing Union citizen has taken place within the Schengen area, where internal borders may be crossed without checks being carried out on individuals. At the same time, he seems to assume that EU citizens still have to “apply” for a right of residence. He even states that the FS case concerns circumstances in which an EU citizen is relying on a new right of residence in a Member State from which s/he was expelled “without in the meantime having been formally granted a right of residence in another Member State”.
These statements are problematic in light of EU law. The thorny issue is of course the constitutionalised status of EU citizenship and free movement rights. According to a long line of case law, dating back to the 1976 Royer case, free movement rights are exercised directly on the basis of the Treaties and independently from national administrative action. The result is that Union law does not acknowledge such a thing as a “formal” right of residence for EU citizens and all types of formalistic bureaucratic procedures or paperwork are simply seen as “declaratory” not as “constitutive” of free movement rights. This was codified, although somewhat ambiguously, in the Citizenship Directive, where art. 8 allows Member States to register Union citizens and verify their right of residence after three months of residence but art. 25 prohibits Member States from making any administrative formality a “precondition” for exercising free movement rights under the Treaty. The result is that that the actual residence status of many EU citizens remains undetermined, unknown and highly fluid as it exists independently from official paperwork and registration databases; a constellation that might have empowered EU citizens in the exercise of their free movement rights against Member State bureaucracies but also added to confusion and uncertainty in national authorities and EU citizens themselves.
Over the past decade, an increasing number of Member States have responded to this constitutionalisation of free movement rights by adjusting their bureaucratic machineries of dealing with EU citizens. Some have actually given up on exercising any kind of prospective control on whether newly arriving EU citizens comply (or are likely to comply) with the residence conditions. This makes the problem of how to deal with the return of expelled EU citizens even much more profound than how the AG presents it. The two Member States that are relevant in this case for example, the Netherlands and Germany, abolished a duty for EU citizens to register and obtain a registration certificate in 2014 and 2013 respectively. EU citizens arriving on their territories can no longer “apply” for a document certifying their right of residence and their compliance with residence conditions is only verified retrospectively, at the moment they apply for social assistance or a permanent right of residence after 5 years. The solution sketched by the AG, of “applying for a new right of residence” is not readily available in these states. Although this might be surmountable by (re)opening an administrative office, the more difficult issue (from a point of law) is that this formalistic step might be at odds with the doctrine that EU citizens should be able to exercise free movement rights (like a non-discriminatory right to social benefits) irrespective of this bureaucratic obstacle.
Taking the earlier jurisprudence of the Court (and the content of the Citizenship Directive) seriously means that a (renewed) right of residence cannot be made dependent on an obligatory visit to an Immigration office although this might be the clearest and most effective solution. It should arise somehow “automatically” – directly on the basis of the Treaties – when an EU citizen re-enters or substantively changes his/her situation or activities, enabling the EU citizen to enjoy substantive rights and engage with authorities without first having to organise the necessary paperwork or asking officials to change a code in a database.
Member States are allowed to expel Union citizens for their poverty, but the effectiveness of those expulsions is heavily diminished when those citizens are able to return in a context of open borders. This Insight critically discussed the opinion by Advocate-General Rantos in the FS case, where he proposed to require Member States to conduct an “individual examination” as a solution to this problem. It was argued that his solution, although especially the line of reasoning leading up to it, is situated awkwardly within existing legal doctrine and should have shown more awareness of the practical functioning of Member State bureaucracies. A slimmed down version of the Dutch proposal would not be a bad alternative. Three months seems a reasonable period to ask someone to leave the country after which the assumption should be made that someone has complied with the expulsion decision and may take up a renewed right of residence under art. 6 of the Directive. Of course, EU citizens should always be able to return when there is a material change in their residence status (e.g., by finding work or by having a genuine chance of finding work) or return temporarily when concrete reasons (like the court hearing in the current case) so require, but three months of absence could function as a presumptive threshold for satisfying an expulsion decision. It would provide legal clarity and certainty to both administrative authorities and EU citizens and would avoid adding another “case-by-case” assessment to the ever inflating use of individual examinations that make up the EU acquis. Especially the suggestion to make national authorities assess the “risk” that someone will become an unreasonable burden on the social assistance system would in my opinion allow for too much discretion on the side of Member States to keep undesired EU citizens out and might function as a (forbidden) re-entry ban in practice.
* Assistant Professor, Department of Transnational Legal Studies, Vrije Universiteit (Amsterdam), email@example.com. I am grateful to the constructive comments from the anonymous reviewer. Any remaining errors are my own.
 Directive 2004/38/EC of the European Parliament and of the Council of 29 April 2004 on the right of citizens of the Union and their family members to move and reside freely within the territory of the Member States (“Citizenship Directive”).
 Case C‑719/19 FS ECLI:EU:C:2021:104, opinion of AG Santos (“Opinion”).
 The official request for a preliminary ruling mentions suspicions of shoplifting and pickpocketing, as well as arrests for not being able to show his ID-card, trespassing a home and violence against a civil servant. Judgment for referral, Council of State, 201809965/1/V3 (25 September 2019).
 Opinion of AG Santos cit. paras 14-19.
 Ibid. para. 26.
 Ibid. para. 27.
 Ibid. para. 29.
 Case C‑94/18 Chenchooliah ECLI:EU:C:2019:693 para. 74.
 Opinion of AG Santos cit. paras 50-51. The AG bases himself on recital 16 and art. 14(3) of the Directive.
 Opinion of AG Santos cit. para. 72.
 Ibid. paras 74-76.
 Ibid. para. 93.
 Ibid. para. 89
 Ibid. paras 90-91.
 Ibid. para. 92.
 Ibid. para. 98.
 Ibid. para. 100. Intuitively, it feels weird to impose higher return conditions on someone who has lived longer in the Member State and actually made an effort to settle and integrate than from someone who has lived shorter in that Member State and has not made such an effort.
 Ibid. para. 101.
 Ibid. para. 102.
 Ibid. para. 34.
 These mostly concern persons without a permanent address and who cause nuisances, D Kramer, ‘‘In Search of the Law’: Governing Homeless EU Citizens in a State of Legal Ambiguity’ (2017) ACCESS EUROPE Research Paper.
 Opinion of AG Santos cit. para. 56 and dictum.
 Case C-140/12 Brey ECLI:EU:C:2013:565 para. 61.
 Only slightly changing the phrasing in the English language version to “all assistance schemes established by the public authorities”. Case C-333/13 Dano ECLI:EU:C:2014:2358 para. 63; case C-67/14 Alimanovic EU:C:2015:597 para. 44.
 Along the lines of case C-209/03 Bidar ECLI:EU:C:2005:169 para. 56.
 The result would be a true “catch-22” as those EU citizens would because of their legal residence be entitled to social assistance benefits as long as they do not get it. G Davies, ‘Migrant Union Citizens and Social Assistance: Trying to Be Reasonable About Self-Sufficiency’ (2016) College of Europe Research Papers in Law, 13.
 Opinion of AG Wathelet case C‑442/16 Gusa ECLI:EU:C:2017:607 para. 34.
 Case C 710/19 G.M.A. ECLI:EU:C:2020:1037 para. 42.
 This kind of reasoning can best be found in Alimanovic cit. para. 61.
 By offering them a procedural right to be “assessed at all”, C O'Brien, ‘Real Links, Abstract Rights and False Alarms: the Relationship Between the ECJ's "Real Link" Case Law and National Solidarity’ (2008) ELR 643, 650-656.
 M Blauberger and SK Schmidt, ‘Free movement, the welfare state, and the European Union's over-constitutionalization: Administrating contradictions’ (2017) Public Administration 437.
 Opinion of AG Santos cit. para. 92.
 Ibid. para. 83.
 Ibid. para. 34.
 Ibid. para. 58.
 Case C-48/75 Royer ECLI:EU:C:1976:57 paras 31-32. See also case C-85/96 Martínez Sala ECLI:EU:C:1998:217 para. 53.
 Probably clearest are paras 47, 48 and 54 of case C-325/09 Dias EU:C:2011:498. Occasionally, the Court attributed some weight to the fact that Member States had granted a “formal” right of residence, e.g., case C-456/02 Trojani EU:C:2004:488 para. 37.
 Recital 11 is more explicit by stating that “the fundamental and personal right of residence in another Member State is conferred directly on Union citizens by the Treaty and is not dependent upon their having fulfilled administrative procedures”.
 We analyse and theorise this process and the responses in different Member States, including the Netherlands and Germany, in D Kramer and A Heindlmaier, ‘Administering the Union Citizen in Need: Between Welfare State Bureaucracy and Migration Control’ (2021) Journal of European Social Policy, journals.sagepub.com.
 See already G Davies, ‘Bureaucracy and Free Movement: A Conflict of Form and Substance’ (2003) Nederlands Tijdschrift voor Europees Recht, 81.