Coordination of social security under the Withdrawal Agreement

In this post, Professor Tamara Hervey, Natalia Miernik and James Murphy (each of  the University of Sheffield) examine the social security provisions of the Withdrawal Agreement between the UK and the EU.

Introduction

Coordination of social security within the European Union’s internal market is a crucial factor in making free movement of people a practical reality. Without it, people who migrate around the EU during their working and retired lives would lose out on acquired social security entitlements that they would otherwise earn if they stayed within one country. The EU’s scheme and system (which might be described, simply, as a portable ‘backpack’ of earned benefit entitlements) is far more advanced and complex than any other equivalent system maintained through a bilateral social security coordination agreement, although these agreements do exist. The administrative structures of the highly complex EU social security coordination system are also vastly superior to any bilateral arrangements, and the transparency to human beings of their individual entitlements in EU law equally so.

One of the reasons that coordination of EU social security law is complex is because it is a blend of domestic and EU law. Competence to provide benefits in the event of illness, invalidity, old-age, or unemployment rests at national (and sometimes sub-national) level in Europe. But EU law mandates a cooperation between nationally-determined social security systems. Unlike in areas of EU law where harmonisation is the aim, in EU social security law key legal concepts (such as ‘residence’ or ‘employment’) are nationally defined, not a matter of EU law. This blend of the domestic and the EU-level makes determining specific legal entitlements in particular situations unusually challenging. That challenge is perpetuated under the Withdrawal Agreement. In addition, questions about enforceability of the Withdrawal Agreement add a layer of precarity.

Unlike residence entitlements, this aspect of the Withdrawal Agreement has, hitherto, been little discussed in either legal or media analyses (a couple of exceptions here and here). Yet in terms of its long-term effects, it is at least as important, if not more so.

The structure of the Withdrawal Agreement’s provisions on social security and scope rules

Part Two of the Withdrawal Agreement covers citizens’ rights. Title I covers general provisions, including definitions, a general non-discrimination clause (Article 12), and a personal scope provision (Article 10). Title II, entitled ‘rights and obligations’ covers residence rights, rights of workers and self-employed persons, and recognition of professional qualifications. Title III, entitled ‘coordination of social security systems’, in fact also grants rights and imposes obligations. It also has its own scope provision (Article 30), which cross-refers to Article 10, providing that Union citizens, UK nationals, and their family members and survivors who do not, or no longer, fall within Article 30, but who do fall within Article 10, are covered by Title III.

Title IV includes the provision (Article 39) that people covered by Part Two of the Withdrawal Agreement have life-long protection under the Withdrawal Agreement, so long as they continue to meet the conditions set out in either or both of Title II or Title III. Their rights continue beyond the end of the transition period, and irrespective of any future (trade) agreements between the EU and the UK.

It is obviously essential to determine whether someone falls within the scope of the Withdrawal Agreement in order to determine their rights under it. The scope rules of Title II and Title III are different. Residence rights intersect with access to social security, not least because many social security rights are contingent on being ordinarily resident within a particular jurisdiction. Ordinary residence in a social security sense is a matter of domestic law, whereas residence rights in an immigration sense are a matter of Withdrawal Agreement law. Yet, in practice, the two types of rights overlap and intersect. Further, someone may fall within the scope of Title III, even if they do not fall within the scope of Title II. The structure of Part Two of the Withdrawal Agreement, and the different scope rules, add a level of complexity which is probably unavoidable, given the very different nature of the rights and obligations at issue.

The scope rules for Title II are relatively well-understood and have been quite widely discussed. The scope rules for Title III less so.

The starting point is not residence, but being ‘subject to’ the social security legislation of an EU Member State/the UK. Title III applies to ‘United Kingdom nationals who are subject to the legislation of a Member State at the end of the transition period, as well as their family members and survivors’ (Article 30 (1) (b)). Title III also applies to ‘United Kingdom nationals who reside in a Member State, and are subject to the legislation of the United Kingdom at the end of the transition period, as well as their family members and survivors’ (Article 30(1)(d)) and ‘United Kingdom nationals who pursue an activity as an employed or self-employed person in one or more Member States at the end of the transition period, and who, based on Title II of Regulation (EC) No 883/2004, are subject to the legislation of the United Kingdom, as well as their family members and survivors’ (Article 30(1)(e)). People falling within the scope of Title III do so only as long as they continue without interruption to be ‘in one of the situations … involving both a Member State and the United Kingdom at the same time’ (Article 30(2)).

Being ‘subject to’ the relevant social security legislation has not been formally defined in EU law, although Regulation 883/2004, Articles 11-16 provides rules for determining which single state is the responsible state. It is not clear, for example, whether someone who claims they are subject to national social security law, where the national social security authorities deny this, would fall within the scope of the Withdrawal Agreement.

Moreover, people who fall within the scope rules of Article 10, and their families and survivors, also fall within Title III (Article 30 (3)). These rules are based on residence rights having been exercised before the end of the transition period. This provision only covers people for as long as they continue to have a right to reside (under Article 13) or to work (under Article 24 or 25) in the host state. Articles 14 to 23 provide further detail on the necessary conditions for residence, including the right of permanent residence (Article 15), issuance of residence documents (Article 18) and restrictions of the right of residence and entry (Article 20).

The Withdrawal Agreement’s scope rules are of critical importance, as once someone falls outside the scope of the Agreement, they lose all their entitlements and protections under it forever. Once lost, rights under the Agreement cannot be regained.

Lastly, we note that the Withdrawal Agreement does not explicitly provide for the position of bilateral social security agreements. However, the Agreement provides that it ‘shall not affect’ domestic law which is more favourable to the beneficiary. In our view, this should be interpreted as meaning that the Withdrawal Agreement applies in preference to any bilateral social security agreements, except where the bilateral agreement is more favourable for the beneficiary. This interpretation would be consistent with the position in EU law.

Rights and obligations

The provisions for coordination of social security systems are found in Title III. If a UK citizen falls within the scope of Title III, Article 31 provides that the social security coordination rules and objectives of EU law (Article 48 TFEU, Regulations 883/2004 and 987/2009) ‘shall apply’. These rules include entitlements to access to a range of social security entitlements, including unemployment benefits, healthcare and pension entitlements. The objectives of these rules are to secure equivalence for migrant workers and their families to the protections that European welfare states provide in the event of the life events of unemployment, illness, old age and so on, that mean someone needs to rely on the social welfare provisions of a state.

What is covered

The full range of provisions of EU social security coordination law are extended to those within the scope of Title III of the Withdrawal Agreement. These include, for example, the general principle of non-discrimination (Article 31(1) by reference to Regulation 883/2004, Article 4); access to pensions (including the principles of exportability and aggregation) (Article 31(1) by reference to Regulation 883/2004, Articles 3(1)(d), 6, 7, 50ff); access to unemployment benefit (Article 31(1) by reference to Regulation 883/2004, Articles 3(1) (d), 6, 7, 61ff); and access to healthcare (Article 31(1) by reference to Regulation 883/2004, Article 3(1)(a), 7, 17ff).

What is not covered

The major gaps in coverage concern the end of the transition period, and the provision to the effect that, once lost, rights cannot be regained, effectively precluding retention of rights after ‘onward movement’.  Obviously someone moving from the EU to the UK or vice versa after 31 December 2020 cannot acquire residence rights under the Withdrawal Agreement, and so will not fall within the scope of Title III via Article 30 (3)/10. Further, entitlements to residence status are not retained under the Withdrawal Agreement if someone resides outside of the host state for more than 6 months (or 5 years for permanent residents). There is no access to healthcare under the Withdrawal Agreement during a temporary stay in another Member State/UK if the person seeking healthcare arrived in the EU/UK on a visit after 31 December 2020.

By way of illustration, imagine a UK citizen wishing to move to and retire in Spain in January 2021. She would not fall within the personal scope of Part Two of the Withdrawal Agreement and therefore would not be able to benefit from the residence rights therein. Her rights to enter and reside in Spain would solely depend on Spanish national law. She would not be ‘subject to’ Spanish social security legislation at the end of the transition period (she would be ‘subject to’ UK social security law), and thus would fall outside of the scope of the Withdrawal Agreement.

That is a simple example. A slightly more complex one is a UK citizen who has lived and worked in France for 6 years before the end of the transition period. Under Article 15(1) that person has gained permanent residence status and will continue to be a permanent resident of France unless they leave France for ‘period exceeding 5 continuous years’ (Article 15(3)). Therefore, if the UK citizen decides to move to a different Member State to retire, Italy for example, they would lose their permanent residence in France if they did not return to France within 5 years. In terms of the UK citizen’s pension entitlements, they are ‘subject to’ the French social security system, and, as they retain their permanent residence right in France for the first five years of moving to Italy, they therefore fall within the scope of the Withdrawal Agreement by virtue of Article 30(3)  coupled with Article 10(1)(b). This means that the UK citizen will benefit from the pension benefits required by the Withdrawal Agreement, including the principle of aggregation of periods. However, it is far from clear whether the principle of exportability applies, given that the scope rules mean that the Withdrawal Agreement only applies so long as the person concerned formally continues to be resident in France (irrespective of their actual residence in Italy). What is clear is that, after five years of the UK citizen living in Italy, they will no longer fall under the Withdrawal Agreement and their access to their pension will be solely based on French national law, not the Withdrawal Agreement.

The position of frontier workers is more complex still. Imagine a UK citizen who is living in France and working in Luxembourg in December 2020. Frontier workers are regarded as ‘socially insured’ (i.e. ‘subject to the social security legislation of’) the state in which they work. But they are entitled to access health care in both that state and the state in which they reside, which in effect gives them access to the comprehensive sickness insurance necessary for continued residence in that state, should they cease work, for example because of illness or accident. The imagined UK citizen would have a right to continue to reside in France under the Withdrawal Agreement. If they became unemployed or unable to work they would retain their rights to equal treatment with a Luxembourg national who became unemployed or unable to work, and thus access any unemployment or sickness benefits available to Luxembourg nationals. So long as this was sufficient to prevent them becoming a burden on the French social assistance system, this would mean they were entitled to continue to reside in France.

But imagine that this frontier worker, while still resident in France, then took up a job in Germany, after 1 January 2021. There would be no entitlement to do so under the Withdrawal Agreement, but they could do so on the basis of German law. They would however remain within the scope of Title III of the Withdrawal Agreement, as Article 10 (1) covers ‘UK nationals who exercised their right to reside in a Member State’ before the end of 2020.

As someone within the scope of Title III, this UK national frontier worker would be entitled to social security benefits, such as healthcare, in the same way as if they had remained resident in France and working in Luxembourg (the position before the end of 2020). But here is where the interaction between the scope rules in Title III and those elsewhere in the Withdrawal Agreement becomes difficult to understand. This is because the rules in Title III derive from the ‘single state rule’: a principle of EU social security law to the effect that an individual’s social security entitlements flow from a relationship with a single ‘competent state’. It is far from clear from the wording of the Withdrawal Agreement which is the competent state in this context.

There are three possible approaches to interpretation of the relevant provisions:

1. since this frontier worker’s relationship with the competent state is not based on the Withdrawal Agreement in terms of their employment, the coordination rules cease to apply;
2. the general coordination rules apply, including those that determine the competent state, even though the worker’s entitlement to work in the competent state is not based on the Withdrawal Agreement; and
3. the exceptional coordination rule in Article 11 (3) (e) of Regulation 883/2004 applies to the effect that the state of residence is the competent state.

If the first interpretation is correct, the imagined frontier worker will be employed in Germany solely under German national law and not under the WA, meaning that Germany cannot be a competent Member State for any purpose. Therefore, as there is no competent state to satisfy the requirements of Article 17 of the Regulation, then the worker will not be able to benefit from the sickness benefits that Article 17 would confer on them in France had there been a competent state. France is not the competent state, as the worker is a frontier worker, and Regulation 883/2004 provides that, for frontier workers, the competent state is the state of work (Article 11 (3)(a), Regulation 883/2004). The worker would only have entitlements to access to healthcare in France or Germany on the basis of national law, not under the Withdrawal Agreement.

If the second interpretation is correct, then, although the worker has no rights to work in Germany under the WA, Germany is nonetheless to be regarded as the competent state for the purposes of Regulation 883/2004 (Articles 11 (3) (a), 17 & 18), since George falls within the scope of Title III of the Withdrawal Agreement with regards to social security coordination. Accordingly, the worker, who resides in France, a Member State other than the competent Member State (Germany), will receive benefits, including sickness benefits (Article 3 (1)(a), Regulation No 883/2004), in France as though they were insured under the French legislation. This will be provided for on behalf of the competent institution, Germany, by the institution of the place of residence i.e. France (Article 17, Regulation No 883/2004).

If the second interpretation is correct then the worker will also be entitled to access to healthcare while in Germany, the place of work, under Regulation 883/2004, Article 18. This will be provided for by Germany at its own expense, in accordance with the legislation of Germany, as though the worker were a resident of Germany (Article 18 (1), Regulation No 883/2004).

If the third interpretation is correct, rather than the general rule for determination of the competent state (Article 11 (3)(a), Regulation 883/2004), the exception in Article 11 (3) (e) of Regulation 883/2004 applies. Article 11 (3) (e) provides:

‘any other person to whom subparagraphs (a) to (d) do not apply shall be subject to the Member State of residence, without prejudice to other provisions of this Regulation guaranteeing him benefits under the legislation of one or more other Member States.’

The imagined worker is ‘pursuing an activity as an employed person in a Member State’ (Article 11 (3) (a)), so taken literally, does not fall within Article 11 (3) (e). However, it could be argued that the worker is not pursuing that activity as an employed person in the sense of EU law, therefore Article 11 (3) (a) does not apply. If this interpretation is correct, the worker will fall outside of the provisions in Article 17 and 18 Regulation 883/2004, as they will be resident in the competent state. They will be entitled to receive healthcare in France, but not in Germany, except necessary healthcare during the course of a stay in Germany (Article 19, Regulation 883/2004). France will be responsible for paying for healthcare provided in France or (if necessary healthcare) in Germany. This interpretation is consistent, by analogy, with the Commission’s interpretation of the rules in its Guidance Note, p 38.

The situation becomes even more complex when we consider the rights of this imagined frontier worker after they have retired.

In principle, the competent state for the purposes of pension provision is usually the state in which the person last worked. This principle follows from the general rules in Regulation 883/2004, to the effect that the ‘competent institution’ is the ‘institution in which the person concerned is insured at the time of the application for the benefit’ (Article 1 (q), Regulation 883/2004). Under EU law, when someone retires and seeks a pension, they do so from the state in which they last worked, as the place where they are insured at the date of retirement. In this instance, that would have been Germany, under ordinary EU law.

But which is the competent state for a frontier worker like the one we are imagining under the Withdrawal Agreement? As above, the question as to whether, and if so how, the principle of aggregation under the Withdrawal Agreement applies to frontier workers who move their place of work from one Member State to another post-transition, is moot. We again consider four possible interpretations:

1. since the worker’s relationship with Germany is not based on the WA, the coordination rules cease to apply.
2. the general coordination rules apply, including those that determine the competent state, even though the entitlement to work in Germany is not based on the WA.
3. the general coordination rules apply, but to the effect that the last state in which the worker worked under the WA (Luxembourg) is the competent state.
4. the exceptional coordination rule in Article 11 (3) (e) of Regulation 883/2004 applies to the effect that the state of residence (France) is the competent state.

The first interpretation is the most punitive. It would mean that, when they apply to Germany to receive a state pension, this person’s rights would be governed only by German national law. Germany would have no obligation to take into account their previous periods of employment in the UK or Luxembourg and accumulate them. Hence, only employment in Germany would be taken into account and in that case, unless they worked in Germany for more than 5 years, this would not satisfy the minimum 5 year employment period to qualify for a state pension in Germany. Therefore, under this interpretation, the frontier worker would not receive a state pension for their employment in Germany at all.

If this interpretation is correct, the worker would have no entitlements under the Withdrawal Agreement to access their pension accrued both before and after the transition period. They would have to rely on any bilateral agreements between the relevant states. This interpretation is contrary to the spirit of the Withdrawal Agreement, Part Two, which seeks to secure ‘reciprocal protection for Union citizens and for UK nationals … where they had exercised free movement rights before a date set in this Agreement, … recognising also that rights deriving from periods of social security insurance should be protected’ (Recital 6). For this reason, this is unlikely to be the correct interpretation of the terms of the Withdrawal Agreement.

If the second interpretation is correct then this means that pension entitlements would be governed by Germany’s competent institution, Germany being the last place of work. Germany would be obliged to take into account, under Article 6 of the Regulation, the pension rights the worker accumulated in all their places of employment across the EU and UK. Under this interpretation, the competent state for the purposes of the pension would be Germany, even though the worker did not work in Germany under an entitlement granted by the Withdrawal Agreement. This is a logical consequence of the scope rules for Title III being different from those for Title II, and the application of EU law’s system of coordination of social security by the Withdrawal Agreement to those falling within the scope of Title III.

It is arguable that Luxembourg is the ‘competent state’ for the pension entitlements under the Withdrawal Agreement, because Luxembourg is the last state in which the worker worked in reliance on the Withdrawal Agreement. Under Regulation 883/2004, Article 6, Luxembourg would have an obligation to aggregate the period of employment in Luxembourg with previous periods of employment in the UK and elsewhere in the EU. Luxembourg would have no obligation under the Withdrawal Agreement to include the period of employment in Germany. (We do not consider the possible application of bilateral agreements in this context, noting only that Article 8, Regulation 883/2004 provides that any more favourable bilateral agreements shall apply instead of the Regulation.)

Finally, as noted above, it is possible under Regulation 883/2004, Article 11 (3)(e) that France is the competent state. This would be an extraordinary interpretation, given that the worker would never have worked in France, and so has not accumulated pension entitlements in France at all.

We write here of imaginary examples, but of course there are actually many people whose lives are such that their real legal entitlements will be covered (or not) by the Withdrawal Agreement. These are just a couple of examples where the meaning and significance of the Withdrawal Agreement’s Title III provisions are not clear. Further examples can be found in our briefing paper, available from the Health Governance after Brexit website.

Conclusion

The provisions of the Part Two of the Withdrawal Agreement are about human beings: our lives, livelihoods and the protections that European states provide for us when we are or become unable to work, when we retire and when we need healthcare. Although the Withdrawal Agreement’s provisions will continue to apply long after 31 December 2020, their coverage will not be complete, in terms of the way that real people actually live, between the EU and the UK. The gaps in coverage in the Withdrawal Agreement could of course be alleviated, or even nearly eliminated, by one or more agreements between the EU and the UK. At time of writing, however, unfortunately, such an agreement or agreements have not been reached.

The support of the ESRC’s Health Governance After Brexit grant (ES/S00730X/1) is gratefully acknowledged.  This analysis must not be treated as formal legal advice, not least because its authors are not insured to give such advice. Anyone seeking such advice should consult a solicitor or public-access qualified barrister.

Share this post on social media: