The final UK renegotiation deal: immigration issues

Steve Peers*

So David Cameron has achieved his deal on the renegotiation of the UK’s EU membership (full text of that deal here). This is the first of a series of posts on the final deal – starting with the issue of ‘EU immigration’ (or, from the EU law point of view, the free movement of EU citizens). This builds on (and partly recycles) my earlier post on the EU immigration issues in the draft deal.

I will write later about the other substantive issues (competitiveness, Eurozone relations, sovereignty) and on the legal form of the deal (although see already my post on the legal form of the draft deal; my comments there won’t change much when I update them in light of the final deal). And see also Katarzyna Granat's analysis of the 'red card' for national parliaments – again, the final text of the deal doesn’t differ from the draft here).

The deal takes the form of seven legal texts: a Decision of the EU Member States’ Heads of State and Government (the ‘Decision’); a Statement of the Heads of State and Government (which consists of an agreed Council Decision); a Declaration by the European Council (which consists of the EU Member States’ Heads of State and Government, although when acting collectively they are legally distinct from the European Council): and four declarations by the Commission. Of these, Section D of the draft Decision and three of the Commission declarations relate to immigration issues. One of these Commission declarations (relating to child benefit exports) was added during the negotiation, while the text of Section D and another declaration (on the ‘emergency brake’ in in-work benefits) was amended. The other declaration (on so-called ‘abuse’ of free movement) was not changed.

While Section D contains some important attempts to clarify EU free movement law, the key feature of the deal on immigration is the intention to propose amendments to the three main current EU laws. These three laws are: (a) the EU citizens’ Directive, which sets out the main rules on most EU citizens moving to other Member States: (b) the EU Regulation on free movement of workers, which contains some specific rules on workers who move; and (c) the Regulation on social security, which sets out rules on coordination and equal treatment in social security for those who move between Member States. 

All three sets of amendments are to be proposed by the Commission as soon as the main Decision enters into effect. That will happen (see Section E of the Decision) as soon as the UK announces that it will remain a member of the EU – if, of course, the UK public vote to remain in the upcoming referendum. The deal includes a commitment from the Commission to make these proposals, and from the other Member States to support their adoption in the EU Council (oddly, the latter commitment does not apply to the planned amendment to the citizens’ Directive, since that proposal is not referred to in the main Decision).

However, all three proposals will be subject to the ‘ordinary legislative procedure’, meaning that they have to be agreed with the European Parliament. It is also possible that their legality would be challenged before the EU Court of Justice. I can’t appraise the political likelihood of the European Parliament approving the proposals, although the largest party (the European People’s Party, made up essentially of centre-right parties like Angela Merkel’s Christian Democrats) has announced that it supports the renegotiation deal in principle, subject to examination of the details. However, I offer some thoughts below about possible challenges to the legality of these laws if they are adopted.

Unlike some other parts of the deal (on the position of non-Eurozone states, and the exemption of the UK from ‘ever closer union’), there is no mention of future Treaty amendments to give effect to any part of the text dealing with free movement (immigration) issues. So the main impact of the deal in this area will come from the three legislative proposals, once adopted. Since those proposals will not be tabled or agreed until after the UK ‘Remain’ vote (if there is one), this means that the analysis of the details is necessarily somewhat speculative. There are some important points of detail that will only be clear once the legislation is proposed and approved. I flag up some of those finer points below.

Although the press discussion has focussed on the ‘emergency brake’ in in-work benefits, there are three categories of issues: benefits (including a couple of points besides that emergency brake); the family members of non-EU citizens; and EU citizens who commit criminal offences. I refer back to Cameron’s November 2014 speech on EU immigration issues (which I analysed here) where relevant.

It should be noted that there is no text in the deal on two of the issues which Cameron had raised: removal of job-seekers if they do not find a job within six months, and a requirement to have a job offer before entry. Both these changes would have required a Treaty amendment, in light of the Antonissen judgment of the CJEU.

Benefits

There are three benefits issues in the draft deal: (a) the ‘emergency brake’ for in-work benefits; (b) the export of child benefit; and (c) benefits for those out of work.

‘Emergency brake’ on in-work benefits

Cameron had called for no access to tax credits, housing benefits and social housing for four years for EU citizens, but later signalled his willingness to compromise on this point. The position of non-workers and job-seekers is discussed below; but the position of workers is legally and politically difficult, since the Treaty guarantees them non-discrimination.

In the end, the deal provides not for permanent discrimination on this issue, but temporary discrimination on the basis of an ‘emergency brake’. The Commission will propose legislation on this issue, which will provide that the UK (or other Member States) can apply a four-year ban on in-work benefits, subject to substantive and procedural criteria. Procedurally, the rules will say that a Member State will apply to the Council to authorise the ban. The Council will presumably act by the default voting rule in the Treaties: a qualified majority on a proposal from the Commission. That means no single Member State can veto the request to pull the brake. The final deal leaves vague the exact authorisation process which will apply in the Council, to avoid annoying the European Parliament (EP); but that detail will have to be addressed sooner or later. Certainly the EP will have to approve the legislation which sets up that process in the first place; the question is whether it would have a role deciding if the brake should be pulled.

A Commission declaration states the UK qualifies to pull this ban immediately, in particular because it did not apply transitional controls to workers from new Member States in 2004. However, there is nothing in the deal to suggest that Member States – who would have the final word – also agree. The restrictions would only to those who were ‘newly arriving for a period of seven years’, and would have to be phased out during that time. Again, the seven years matches the transitional period which the UK could have applied to control the numbers of workers from new Member States, back in 2004.

Several points of detail arise. First of all, after the seven years have expired, it’s not clear how much time would then have to pass before the brake could be applied again. Secondly, it will be important to clarify the meaning of those who are ‘newly arriving’. What about those who lived in the UK before, and are now returning here? How much time would they have had to spend in Poland (say) before they are considered ‘newly arriving’ again? Presumably the brake would not apply to those who are already here when the brake is pulled, but are not working at that time (due to youth, unemployment, childcare or illness) but who get work afterward.  

Thirdly, it will be necessary to define how to calculate the four year period. It’s easy enough to apply it to those who begin work as soon as they (newly) arrive in the country, and who work for the full four years afterward. But what about those (a non-working spouse, or a teenager, for instance) who start work some time after they enter the country? What about those who start work, stop for whatever reason and then restart? What about those who start work during the brake period, then spend a year or so in Poland, then come back? And how can we be sure when exactly someone entered the country in the first place?

The final crucial point of detail is, obviously, the grounds on which the brake can be applied. According to the Decision, it would apply where:

‘an exceptional situation exists on a scale that affects essential aspects of [a Member State’s] social security system, including the primary purpose of its in-work benefits system, or which leads to difficulties which are serious and liable to persist in its employment market or are putting an excessive pressure on the proper functioning of its public services’.

There’s certainly a widespread perception that one of more of these problems exist in the UK and are caused by the large increase in the number of workers from other Member States in recent years. However, there are two serious problems with the proposed mechanism. Firstly, as Jonathan Portes has argued, objective evidence for this view is lacking. Secondly, while the CJEU has been willing to accept certain limits to free movement rights on the grounds of protecting health systems (see my prior blog post for details), this would have a much more far-reaching impact on non-discrimination for workers. It’s certainly conceivable that by analogy from the Court’s obvious willingness to keep EU monetary union afloat, along with its endorsement of restrictions for non-workers in recent years (see below), it might accept that these plans do not violate the Treaties. But as EU law currently stands, that is probably a long shot.

Export of child benefit

Cameron sought to end payment of child benefit to children living in other Member States. This payment is provided for in the EU social security coordination Regulation, which would have to be amended to change those rules. There was a strong argument that the plan would have breached the Treaties, since in the case of Pinna the CJEU struck down EU legislation that allowed Member States not to export such benefits at all as a breach of the rules on free movement of workers.

The deal does not go as far as Cameron wanted: instead child benefit can be limited by indexing it to the ‘conditions’ in the receiving State. This will only apply to ‘new claims made by EU workers in the host Member State’; but after 1 January 2020, this ‘may’ be extended to ‘existing claims already exported by EU workers’. This is clarified by the Commission declaration, which states that the ‘conditions’ refers to the ‘standard of living and level of child benefits’ in the child’s State of residence. The transitional rule, and the Commission declaration, were added during negotiations. It’s an open question whether this new law would breach the Treaties, since there is no case law on the point.

Several points of detail arise here. It’s explicit that the new rules will be optional, so Member States can still be more generous if they want to. There’s nothing to limit their application to the UK (although I will refer to the UK and Poland here, purely for the sake of readability). It’s not clear whether the rules will also apply to British citizens who have children in other Member States; arguably the principle of non-discrimination will require that they do. It’s also not clear what happens to ‘mixed’ families of (say) British and Polish parents (or indeed step-parents). Will it depend on which parent is the worker? What if both are workers? What if that changes over time?

The transitional clause also raises issues. The Decision distinguishes between ‘new claims’ and ‘existing claims already exported by EU workers’. Presumably the new law will state a precise date at which claims can be regarded as ‘existing’ (say 1 January 2017). These must be existing exported claims, so if a child moves to Poland after 1 January 2017, or is born after that date and resides in Poland, then child benefits could be reduced, even if the worker is already in the UK. So if my estimated date is correct, anyone who is thinking about having a child, and who wants to avoid the application of these rules, had better get a move on. Perhaps this Easter will be the season of fertility even more than usual.

Finally, it should be noted that a challenge by the Commission to other aspects of UK payment of child benefit to EU citizens is still pending. The non-binding opinion of an Advocate-General argues in favour of the UK in this case (for a critical view, see Charlotte O’Brien’s analysis here). It wouldn’t surprise me if the Commission quietly withdrew this legal challenge. You read that here first.

Benefits for those out of work

Cameron sought to end social assistance for job-seekers. The EU legislation already rules out social assistance for job-seekers, so this reflects the status quo. Although the CJEU has said that job-seekers have a right to access benefits linked to labour market participation, if they have a link already with the labour market in question, it took a narrow view of this rule in the judgment in Alimanovic. Pure benefit tourists (who have never had work in the host State) are not entitled to benefits, according to the judgment in Dano. So the Decision simply reiterates this case law, which has already satisfied Cameron’s main objectives in this field. It should be noted that another judgment by the Court of Justice on EU benefits issues is due next week.

EU citizens’ family members

Under the EU citizens’ Directive, currently EU citizens can bring with them to another Member State their spouse or partner, the children of both (or either) who are under 21 or dependent, and the dependent parents of either. This applies regardless of whether the family members are EU citizens or not. No further conditions are possible, besides the prospect of a refusal of entry (or subsequent expulsion) on grounds of public policy, public security or public health (on which, see below).

In principle EU law does not apply to UK citizens who wish to bring non-EU family members to the UK, so the UK is free to put in place restrictive rules in those cases (which it has done, as regards income requirements and language rules). However, the CJEU has ruled that UK citizens can move to another Member State (the ‘host Member State’) and be joined by non-EU family members there, under the more generous rules in the EU legislation. Then they can move back to the UK (the ‘home Member State’) with their family members, now invoking the free movement rights in the Treaties. This is known in practice (in the UK) as the ‘Surinder Singh route’, because of the name of the case which first established this principle. In 2014, the CJEU clarified two points about this scenario (as discussed by Chiara Berneri here): (a) it was necessary to spend at least three months in the host Member State exercising EU law rights and residing with the family member, before coming back; and (b) the EU citizens’ Directive applied by analogy to govern the situation of UK citizens who return with their family members.

In his 2014 speech, David Cameron announced his desire to end all distinction between EU citizens and UK citizens as regards admission of non-EU family members, by allowing the UK to impose upon the EU citizens the same strict conditions that apply to UK citizens. Since this would have deterred the free movement of those EU citizens who have non-EU family members, there is a good chance that it would have required not just a legislative amendment but a Treaty change.  (Note that according to the CJEU, EU free movement law does not just require the abolition of discrimination between UK and other EU citizens, but also the abolition of non-discriminatory ‘obstacles’ to free movement).

However, the deal does not go this far. The main Decision states that:

‘In accordance with Union law, Member States are able to take action to prevent abuse of rights or fraud, such as the presentation of forged documents, and address cases of contracting or maintaining of marriages of convenience with third country nationals for the purpose of making use of free movement as a route for regularising unlawful stay in a Member State or for bypassing national immigration rules applying to third country nationals.’

The Commission Declaration then states that it will make a proposal to amend the citizens’ Directive:

‘to exclude, from the scope of free movement rights, third country nationals who had no prior lawful residence in a Member State before marrying a Union citizen or who marry a Union citizen only after the Union citizen has established residence in the host Member State. Accordingly, in such cases, the host Member State's immigration law will apply to the third country national.’

That Declaration also states that the Commission will clarify that:

‘Member States can address specific cases of abuse of free movement rights by Union citizens returning to their Member State of nationality with a non-EU family member where residence in the host Member State has not been sufficiently genuine to create or strengthen family life and had the purpose of evading the application of national immigration rules’; and

‘The concept of marriage of convenience - which is not protected under Union law – also covers a marriage which is maintained for the purpose of enjoying a right of residence by a family member who is not a national of a Member State.’

It seems clear that these ‘clarifications’ will not be included in the legislative proposal, since the declaration later concludes (emphasis added):

‘These clarifications will be developed in a Communication providing guidelines on the application of Union law on the free movement of Union citizens.’

Let’s examine the planned legislative amendments, then the guidelines which will provide ‘clarifications’. The amendments will exclude two separate categories of non-EU citizens from the scope of the citizens’ Directive: those who did not have prior lawful residence in a Member State before marrying an EU citizen who has moved to another Member State; and those who marry such an EU citizen after he or she has moved to a Member State. For these people, national immigration law will apply.

The background to this proposal is CJEU case law. In 2003, in the judgment in Akrich, the CJEU ruled that Member States could insist that non-EU family members had previously been lawfully resident in the Member State concerned (previously no such rule appeared to exist). But in 2008, in Metock, the CJEU overturned this ruling and said that a prior legal residence requirement was not allowed.

Several points arise. First, the basic definition: what is lawful residence exactly? Presumably it means more than lawful presence, ie a stay of three months on the basis of a valid visa or visa waiver. But what about ambiguous cases, such as a pending asylum application or appeal? EU legislation says that asylum-seekers can usually stay until the application fails (if it fails), and then during the appeal (subject to some big exceptions). According to the CJEU, the EU’s main rules on irregular migrants therefore don’t apply to asylum-seekers whose application is pending.

Secondly, it’s odd to refer to national law alone, since sometimes EU law governs the admission of non-EU nationals. Even the UK (along with Ireland) is bound by the first-phase EU asylum law, and by the EU/Turkey association agreement. Denmark is bound by the latter treaty. And all other Member States are bound by the second-phase asylum law, along with EU legislation on admission of students and researchers and some categories of labour migrants (the highly-skilled, seasonal workers and intra-corporate transferees).

Thirdly, it’s arguable that the EU principle of non-discrimination applies. That would mean, for instance, that if a German woman already in the UK married her American husband, the UK would have to treat her the same as a British woman in the same situation – but no worse. This would in fact be relevant to every Member State – there’s nothing in this part of the deal that limits its application to the UK. (One important point of detail is whether all Member States would be obliged to apply the new rules on ‘prior lawful residence’ and ‘marriage after entry of the EU citizen’, or whether they could choose to waive one or both of those rules. The EU citizens’ Directive already states that Member States can apply more liberal standards if they wish to).

Finally, the consequences of the rule will need to be clearer in the future legislative amendments. Does the exclusion from the scope of the Directive mean that the family member is excluded forever from the scope of the citizens’ Directive – even if the person concerned is admitted pursuant to national immigration law? That would mean that national immigration law (or EU immigration legislation, in some cases) would continue to govern issues such as the family member’s access to employment or benefits, or subsequent permanent residence. It’s also not clear what happen to children such as the step-child of the EU citizen, or a child that was born to the EU and non-EU citizen couple while living in a third country.

Could this legislative amendment violate the EU Treaties? In its judgment in Metock, the Court referred almost entirely to the wording of the citizens’ Directive. It mainly referred to the Treaties when concluding that the EU had the competence to regulate the status of EU citizens’ third-country national family members. But it also referred to the Treaty objective of creating an ‘internal market’, as well as the ‘serious obstruct[ion]’ to the exercise of freedoms guaranteed by the Treaty, if EU citizens could not lead a ‘normal family life’. It must therefore be concluded that there is some possibility that the revised rules would be invalid for breach of EU free movement law.

Would the amendment violate the EU Charter right to family life? That’s unlikely. While the right to family life is often invoked to prevent expulsions of family members, the case law of the European Court of Human Rights gives great leeway to Member States to refuse admission of family members, on the grounds that the family could always live ‘elsewhere’ – as the CJEU has itself acknowledged (EP v Council). There is some possibility, though, that the CJEU would be reluctant to follow that case law (EP v Council concerns families entirely consisting of non-EU nationals) in the context of free movement: the idea that you could go away and enjoy your family life somewhere else is antithetical to the logic of free movement.

As for the ‘clarifications’ in future guidelines, they will of course not be binding. They first of all refer to cases where an EU citizen has moved to another Member State and come back to the home State. The definition of what constitutes a ‘sufficiently genuine’ move to another country is set out in the case law (three months’ stay with a family member) and mere guidelines cannot overturn this.

It should be noted that the Surinder Singh case law is in any event derived from the Treaty. This line of case law does not accept that such movement between Member States is an ‘evasion’ of national law – as long as free movement rights are genuinely exercised with a family member for a minimum time. The CJEU also usually assumes (see Metock, for instance) that a ‘marriage of convenience’ cannot apply to cases where there is a genuine relationship, even if an immigration advantage is gained. (The Commission has released guidelines already on the ‘marriage of convenience’ concept: see analysis by Alina Tryfonidou here).

Having said that, the planned legislative changes will complicate the plans of people who wish to move to another Member State with their non-EU family and then move back, since national immigration law will apply to their move to the host Member State. It will be important to see how the legislative amendments address the transitional issues of people who have already moved to a host Member State before the new rules apply. Can the home Member State say, possibly based on the Commission’s ‘guidance’ (which might be issued before the new legislation is adopted) that those families must now obtain lawful residence in the host State for the non-EU family member, before the non-EU family member can come to the home State?

Criminality and free movement law

The Treaties allow for the refusal or entry or expulsion of EU citizens on ‘grounds of public policy, public security or public health’. The citizens’ Directive sets out detailed substantive and procedural rules on this issue, which has been the subject of considerable CJEU case law.

What does the renegotiation deal do? First of all, the Decision states that:

‘Member States may also take the necessary restrictive measures to protect themselves against individuals whose personal conduct is likely to represent a genuine and serious threat to public policy or security. In determining whether the conduct of an individual poses a present threat to public policy or security, Member States may take into account past conduct of the individual concerned and the threat may not always need to be imminent. Even in the absence of a previous criminal conviction, Member States may act on preventative grounds, so long as they are specific to the individual concerned.’

To this end, the Commission declaration states that it will:

‘also clarify that Member States may take into account past conduct of an individual in the determination of whether a Union citizen's conduct poses a "present" threat to public policy or security. They may act on grounds of public policy or public security even in the absence of a previous criminal conviction on preventative grounds but specific to the individual concerned. The Commission will also clarify the notions of "serious grounds of public policy or public security" and "imperative grounds of public security" [grounds for expelling people who have resided for longer periods in a host Member State].  Moreover, on the occasion of a future revision of [the citizens’ Directive], the Commission will examine the thresholds to which these notions are connected.’

It’s not clear whether the revision of the Directive referred to at the end here is as imminent as the proposal to amend the rules to create a ‘prior lawful residence’ rule for non-EU family members. Otherwise the plan to issue guidelines is clearly not binding. The language in these guidelines partly reflects the existing law, but some features are new: the greater emphasis on past conduct, the lesser need to show that a threat is imminent and the possibility of expelling someone as a ‘preventative’ measure.

These changes fall within the scope of Cameron’s desire to have ‘stronger measures to deport EU criminals’. However, it should be noted that there is no specific reference to his plans for ‘tougher and longer re-entry bans for foreign rough sleepers, beggars and fraudsters’. While a conviction and re-entry ban for fraud might be covered by the guidelines referred to above, there’s no mention of clarifying entry bans as regards those guidelines, or changing the legislation on this issue. Also, as I noted in my comments on Cameron’s plans at the time, EU legislation does not allow for re-entry bans for rough sleepers and beggars, since the EU citizens’ Directive states unambiguously that a ban on entry cannot be imposed where a person was expelled for grounds other than public policy, public security and public health. Put simply, a Member State can impose an entry ban where an EU citizen has been expelled due to criminality – but not where he or she has been expelled due to poverty.

Longer waiting periods for free movement of persons from new Member States

Finally, it should be noted that the Decision briefly refers to Cameron’s plan to have longer waiting periods for free movement of persons in future accession treaties. It does not incorporate his suggestion, but merely notes it. However, since the details of each new Member State’s adaptation to EU law are set out in each accession treaty, which has to be approved by each Member State, the UK can simply veto any future accession treaties unless longer waiting periods for free movement are indeed included. The next accession to the EU is at least four years away, probably more. So nothing really turns on the absence of agreement with the UK’s position for now.

Conclusion

The key point to remember about the renegotiation deal, particularly as regards EU immigration, is that it consists of different parts. The main deal takes the form of a Decision, which essentially clarifies EU law without amending it. According to CJEU case law (Rottmann), the Court is willing to take Decisions like these into account when interpreting EU law.

However, in the area of EU immigration, the other parts of the deal are more relevant: the intention to pass three new EU secondary laws. Those new laws will be a fully-fledged amendment to existing EU rules, not simply a clarification of it. While some points of detail remain to be worked out, it is clear from the deal that the Commission will make proposals in these areas, and all Member States (ie the Council) will support them. It remains to be seen whether the European Parliament will approve them, and whether the CJEU would accept challenges to their legality. My assessment of the Court’s likely response, as detailed above, is that the amendments on family members will probably be acceptable; the child benefit reforms are an open question; and the changes on in-work benefits are highly vulnerable. Of course, there’s no prior case law on these specific issues, and so we can’t be certain of the Court’s approach in advance.

Overall, as I concluded in the earlier post on the draft agreement, these changes, if they are all implemented as planned, will fall short of a fundamental change in the UK’s relationship with the EU. But equally it is clearly wrong to say that they mean nothing – if in fact they are implemented. The changes would be modest but significant: amendments to three key pieces of EU legislation that would for the first time roll back EU free movement law, not extend it. Leaving aside the calls for non-binding guidelines, there would be cutbacks in in-work benefits (albeit for a limited period), significantly more control on the admission of non-EU family members of EU citizens, and more limited export of child benefit.

Barnard & Peers: chapter 13

Photo credit: www.telegraph.co.uk

*Disclosure: I will be consulting for the European Parliament on the free movement aspects of the renegotiation. However, my advice will be fully independent; I don't represent or advocate for the European Parliament (or anyone else) on these (or any other) issues.