PUBLISHED ON EU LAW ANALYSIS on Wednesday, 7 October 2015
by Charlotte O’Brien,
Senior Lecturer, York Law School
The political message being sent by irate governments to ‘back off’ from national welfare systems’ assumed prerogative to discriminate between home nationals and EU nationals is being received and applied with alacrity by the Court of Justice. The current direction of travel resiles from earlier progressive visions of EU citizenship, and in C-140/12 Brey, C-333/13 Dano and C-67/14 Alimanovic we see that which was once ‘destined to be [our] fundamental status’ receding ever further from view. Advocate General Cruz Villalón’s Opinion in Commission v UK continues the retreat, arguing that the Commission’s action challenging the UK right to reside test for family benefits should be dismissed. The result may, in the current environment, be unsurprising. But getting there with existing legal tools is problematic.
The Opinion contains a number of uncomfortable contortions to give undue deference to the national rules, and avoid tackling the underlying conflict of rules and approaches. It represents quite startling judicial activism in embroidering the legislation with unwritten limitations as to personal scope, tinkering with the subject matter, and asserting an unwritten licence to discriminate whenever something smells like a welfare benefit. The effect is as though the Court’s new teleological guiding principle should be that the legislature would have wanted at all costs to avoid offending the UK government.
The UK right to reside (RTR) test prevents any EU national who does not meet the criteria in Art 7 Directive 2004/38 from receiving Child Benefit or Child Tax Credit, both of which were accepted as being ‘family benefits’, so ‘pure social security’ (rather than special non-contributory benefits in Brey, Dano and Alimanovic) under Regulation 883/2004. The Commission challenged the test’s lawfulness on two grounds – that it imported extra conditions into the ‘habitual residence’ test, to undermine the effects of Regulation 883/2004, and that it is discriminatory since it only applies to non-UK citizens. The AG’s Opinion is remarkable, in its ability to reject both without engaging with either. This analysis deals with four key issues arising from the Opinion: (i) stitching, splicing and embroidering Reg 883/2004; (ii) the ‘inherent’, ‘inevitable’ and ex ante discrimination fudge; (iii) the parallel reality in which the UK does not presume unlawful residence; and (iv) the failure to notice that the UK automatically refuses social assistance to those reliant on ‘sufficient resources’.
(i) stitching, splicing and embroidering Regulation 883/2004
The AG is at some pains to determine whether the ‘right to reside’ test is part of the habitual residence test (HRT), or a separate test added on, suggesting that it is only if it is presented as the former, does the Commission have a case. As the UK government ‘distanced’ itself during proceedings from the combined test approach, and argued that it was a separate test of lawful residence, so the AG commented that the Commission’s case was ‘weakening over the course of the dispute’. Indeed, on the basis that the test was ‘independent’ of the HRT, the AG argued that the first ground should be dismissed. This is perplexing. It seems to be a matter of regulatory semantics whether the RTR is part of the HRT, or is applied as well as the HRT, if the effect – to undermine Regulation 883/2004 – is the same.
For the record, the conclusion that they are separate tests is unconvincing anyway. For all benefits with an official ‘habitual residence test’ the regulations provide that a claimant cannot be habitually resident unless she has the right to reside in the CTA (Income Support (General) Regulations 1987, reg 21AA; Jobseeker Allowance Regulations 1996, reg 85A; Employment and Support Allowance Regulations 2008, reg 70(2); State Pension Credit Regulations, reg 2; see DWP, DMG, 072771). For CB and CTC the terminology is slightly different – the words ‘habitually resident’ are not used, but a person must be treated as being in the UK. And to be treated as being in the UK, you have to have a right to reside (Child Benefit Regulations 2006, Reg 23(4)(a); Tax Credits (Residence) Regulations 2003, Reg. 3(5); CBTM10010 – Residence and immigration: residence – introduction).
However, whether we treat the RTR as part of habitual residence, or as an extra test, the effect in both cases is to add conditions onto the circumstances in which a person is treated as meeting the ‘residence’ criteria of Regulation 883/2004. That Regulation offers a clear, exhaustive list for allocating ‘competence’ of Member States for benefits, providing a residual category for the economically inactive, at Art 11(3)(e) in which the Member State of residence is competent. Once competence has been established, that State is then responsible for the payment of family benefits, subject to the non-discrimination provision.
The scheme of the Regulation is intentionally broader than that of Directive 2004/38 – applying a different personal scope for a start (covering all those who ‘are or have been subject to the legislation of one or more Member States’), and covering pensioners, those between jobs, those who might fall outside of the Dir 2004/38 Article 7(3) retention provisions – essentially, those who should be covered by social security provisions. To apply the right to reside test is to hack down the rationae personae of the Regulation to emulate that of Directive 2004/38 – an approach not endorsed, implied or merited in the Regulation. The AG’s assertion that law should not exist in ‘separate compartments’ as justification for splicing the instruments together and embroidering an extra condition into the Regulation rather too easily ignores the different purposes and scopes of the instruments. Similarly, the different material issues – the restriction of social assistance now embodied in Directive 2004/38, versus award of social security, are inappropriately assimilated. The AG notes, apparently approvingly, the UK’s assertion that ‘the two benefits at issue in the present case have some characteristics of social assistance’. This goes unexamined, and helps form the context in which the different nature of social security, and different subject matter of the Regulation, is effectively ignored. In sum, we have an approach in which if a benefit is a ‘bit like’ social assistance, and a legal instrument is in roughly the same area as Directive 2004/38, then unwritten restrictions kick in.
In the specific case of family benefits, the Regulation’s residual category should provide a guarantee that families do not fall through the cracks and find themselves disentitled to any family benefits, since many Child Benefits are tied to residence. This also serves the ‘bonus’ purpose of protecting children, who are not the agents of migration, and who the legislature and the Court have hitherto taken pains to protect from suffering the penalties of their parents’ choices and/or misfortunes – either out of an interest in child welfare, or as an instrumental way of avoiding disincentives (risks to their children’s welfare) for workers to migrate.
Here it is worth emphasising that when we speak of falling through the cracks, we mostly speak of people who have been working (rather than those who have never worked). The right to reside test results in a strict bifurcation between those ‘working’ and those not. The rules on retention of worker status are stringent and exclusionary, so that people can be working and contributing for many years and still fall over welfare cliff edges. Regulation 883/2004 should offer some protection to their pre-school children in such cases, even where Directive 2004/38 is (according to emerging case law) rather harsher to the parents.
However, in the AG’s approach we can see the Directive, having already been transformed from an instrument to promote free movement into a instrument to prevent benefit tourism (Dano); being promoted to the status of a fundamental principle of limitation, to be (retrospectively) mainstreamed into other (higher) legislative instruments – exerting restrictions that are not there written.
(ii) the ‘inherent’, ‘inevitable’ and ex ante discrimination fudge;
The AG avoided dealing with the question of whether the RTR test discriminates contrary to Regulation 883/2004, by finding that the RTR prevented the Regulation from being applicable at all – apparently treating ex ante discrimination as de facto lawful. This conceptual approach is deeply problematic – can Member States really avoid the non-discrimination obligations contained in legislation by applying discriminatory gateways to access that legislation?
As noted above, once competence of a Member State has been established for the purposes of Regulation 883/2004, it is then – according to that instrument, bound by non-discrimination duties (Article 4). However, under the proposed approach, there will be people for whom no Member State has competence, because competence is to be determined according to a set of restrictions in a completely different instrument which apply a different concept to a different set of people for a different set of benefits. And if they are in this way found not be within any State’s competence, the question of discrimination is avoided.
To the extent that the AG does engage with non-discrimination duties, it is part of an imprecise discussion about the likelihood of the lawfulness of curbing benefits from non-nationals (benefit restrictions are ‘traditionally associated’ with requirements of legal residence). In drawing upon Dano and Brey, the fact that those cases dealt with benefits therein defined as social assistance is swept aside somewhat as the AG finds ‘there is nothing in those judgments to indicate that such findings apply exclusively to the social assistance benefits or the special non-contributory cash benefits with which those cases were concerned and not to other social benefits’. But there is plenty to indicate that social security benefits should be treated differently in their coverage in a different piece of legislation. It is surely very odd to suggest that the Court should list those instruments on which it was not ruling.
Recognising that the rules do treat UK nationals and non nationals differently, the Opinion makes some rhetorical points about discrimination as part of the natural ecosystem of free movement – ‘one way of looking at it is that this difference in treatment as regards the right of residence is inherent in the system and, to a certain extent, inevitable… In other words, the difference in treatment between UK nationals and nationals of other Member States stems from the very nature of the system.’ None of this does anything to address the question of the problem of direct versus indirect discrimination – the latter being rather easier to justify. It almost suggests that some degree of direct discrimination has to be accepted as a matter of pragmatism. Indeed, the characterisation of the rules asindirectly discriminating on the grounds of nationality is one of the most contentious issues in the case. Much as in C-184/99 Grzelczyk, an extra condition is imposed only upon non-nationals. Hiding behind the banner of indirect discrimination seems unconvincing if we posit a brief thought experiment. Imagine all EU national men automatically had an RTR, but all EU national women had to pass the RTR test; that could not be described asindirectly discriminating on the grounds of sex. While it could be argued that nationality is a different type of ground to sex, and so different differences are acceptable, the fact that we are dealing with direct discrimination remains. And this is not explored. The only thing that needs justification, under this analysis, is not the test, but the procedural checking, which we look at next.
(iii) the parallel reality in which the UK does not presume unlawful residence
The AG states that it cannot be inferred that the UK presumes that claimants are unlawfully resident, adding that European citizenship would preclude such a presumption, and that claimants should not systematically be required to prove they are not unlawfully resident.
However, the whole claims process in the UK does systematically require proof of claimants that they are (not un)lawfully resident. The right to reside test takes the limitations of Directive 2004/38 and makes them a priori conditions of the existence of the right to move and reside. There is no general citizenship-based right to reside that can be modified by limitations, with some discretion. The conditions come first, and must be demonstrably met, in each and every case. The UK’s assertion that ‘In cases in which there is doubt as to whether the claimant has a right of residence, an individual assessment of the claimant’s personal circumstances is carried out’ rather masks the process of assessment that decision makers are required to undertake according to the decision maker guidance on establishing whether a claimant really is or was a worker – using the UK’s own definition. That definition is flawed in itself, requiring evidence to meet a higher threshold than set in EU law, and the evidential hurdles can be considerable. Even for the most straightforward cases of worker, proof is required that earnings have been at or above the Minimum Earnings Threshold for a continuous period of at least three months. Those with variable earnings are expected to provide considerable evidence if they wish to ‘prove’ their right to reside. In cases where HMRC have reason to doubt conditions continue to be met for tax credit awards, they issue further, penetrating compliance checks, and in the UK government’s Budget Policy costings document, the government announced that the restrictions on benefits ‘will be augmented by additional HMRC compliance checks to improve detection of when EEA migrants cease to be entitled to these benefits. The checks will apply to all EEA migrant claims’. The system is set up to make the conditions constitutive of the right to free movement, effectively requiring all claimants to prove that they are not unlawfully resident, notwithstanding the apparent ‘background’ of EU citizenship, and claims are subject to systematic checking, notwithstanding Article 14(2) of Directive 2004/38.
The AG however, took the position that such checks are not systematic, but may be indirectly discriminatory, but that they were lawful, with the briefest of nods to justification – as though the mere mention of the UK’s public finances is sufficient to provoke a reverential hush, genuflection and swift retreat from the subject:
‘without any need to pursue the argument further, I consider that the necessity of protecting the host Member State’s public finances, (75) an argument relied on by the United Kingdom, (76) is in principle sufficient justification for a Member State to check the lawfulness of residence at that point.’
No data, it seems, is required.
Nor is any engagement with the question as to whether purely economic aims are legitimate aims for the purpose of justifying discrimination or restricting a fundamental freedom – on this, see AG Sharpston’s Opinion in C-73/08 Bressol.
(iv) the failure to notice that the UK automatically refuses social assistance to those reliant on ‘sufficient resources’.
The AG rounds up the Opinion by noting that in any case, the economically inactive are not completely hung out to dry – they should have their circumstances examined to determine whether they have sufficient resources not to become a burden on the public purse. Here, the AG emphasises that mere recourse to public funds should not bar a claimant from having a right to reside based on sufficient resources, and that their case should be assessed as to whether they are an ‘excessive’ burden. This is all very well, but speaks to a rather different reality to that experienced in the UK, in which the economically inactive are automatically barred from claiming social assistance because they are automatically treated as not having sufficient resources at the point of claim. Moreover, the Upper Tribunal has suggested that ‘sufficient resources’ means sufficient to provide for the migrant’s family for five years; a migrant cannot claim to have had sufficient resources for a short period of time between jobs if those resources would not have lasted for five years.
In short, the Court should be wary of following the AG’s lead in backing off from the apparently prohibited area of UK welfare benefits quite so hastily. The Regulation’s personal and material scope, and purpose, cannot simply be ignored or modified, nor can the Directive be transformed into an all-encompassing, higher principle, through pro-Member State judicial activism. The right to reside test adds conditions to the application of the Regulation’s provisions, and it does so in a directly discriminatory way. The Court must address these points honestly; if it is prevented from doing so by the political wind – or if it too conjures up a default forcefield around benefits regardless of type, and gives licence to ‘inevitable’ discrimination – the ramifications will tell not only upon claimants, their children, the vanishing strands of EU citizenship and the obstructed freedom to move, but also upon the Court’s credibility.