Tuesday, 24 April 2018

Torture victims and EU law

Professor Steve Peers, University of Essex

What happens if an asylum seeker faces severe mental health problems that cannot be treated in the country of origin?  Today’s judgment of the ECJ in the MP case, following a reference from the UK Supreme Court, goes some way towards answering this question.


The issue what we might call “medical cases” for asylum first of all arose before the European Court of Human Rights. In a series of judgments, that Court clarified whether the ban on torture or inhuman or degrading treatment, set out in Article 3 of the European Convention on Human Rights (ECHR), prevented people from being sent back to a country where there was no effective medical care.  Essentially, it ruled that such an argument could only be successful in highly exceptional cases, in particular where the person concerned was critically ill and close to death.

However, while these judgments addressed the question of non-removal for persons in such serious conditions, they did not rule on the issue of the status of asylum, or other types of migration status, for the persons concerned. This issue was the subject of two linked ECJ judgments (M’Bodj and Abdida) in 2014, which I discussed here. In short, the ECJ said that the persons suffering from severe health problems could not invoke a right to “subsidiary protection” on the basis of the EU’s qualification Directive, even though one of the grounds for such protection was facing a “real risk” of torture or inhuman or degrading treatment in the country of origin. That was because subsidiary protection was only intended for cases where the harm was directly caused by humans.

So do “medical cases” only have the right to non-removal on the basis of Article 3 ECHR? Not quite; because the ECJ also said that the EU’s Returns Directive, which governs the position of irregular migrants, could be relevant. In an ambitious interpretation of that Directive, the Court ruled that it could be invoked to prevent removals in “medical cases”, including the suspensive effect of an appeal against removal; moreover the Directive conferred a right to medical care and social assistance for the persons concerned in such cases.

Subsequently, at the end of 2016, the European Court of Human Rights revisited its case law on “medical cases”, lowering the very high threshold that had previously applied before individuals could invoke Article 3 ECHR.  In Paposhvili v Belgium (discussed here), it extended that case law also to cover cases of:

removal of a seriously ill person in which substantial grounds have been shown for believing that he or she, although not at imminent risk of dying, would face a real risk, on account of the absence of appropriate treatment in the receiving country or the lack of access to such treatment, of being exposed to a serious, rapid and irreversible decline in his or her state of health resulting in intense suffering or to a significant reduction in life expectancy.

It should be noted that while the “first phase” EU qualification Directive (which includes the same definition of “subsidiary protection” as the 2011 version) applies to the UK and Ireland, the Returns Directive does not.  


The ECJ began by stating that in order to invoke a claim to subsidiary protection on grounds of torture, it was necessary to show that such treatment would occur in the country of origin in future. While MP had suffered torture in Sri Lanka in the past, that was “not in itself sufficient justification for him to be eligible for subsidiary protection when there is no longer a real risk that such torture will be repeated if he is returned to that country”. Although the qualification Directive states that past serious harm “is a serious indication” there is a real risk of suffering such harm in future, “that does not apply where there are good reasons for believing that the serious harm previously suffered will not be repeated or continue”.

The Court then turned to MP’s health issues, noting that he “presently continues to suffer severe psychological after-effects resulting from the torture” and that “according to duly substantiated medical evidence, those after-effects would be substantially aggravated and lead to a serious risk of him committing suicide if he were returned to his country of origin”. It stated that this provision of the qualification Directive “must be interpreted and applied” consistently with Article 4 of the EU Charter of Fundamental Rights, which set out an “absolute” right to be free from torture or other inhuman or degrading treatment. This Charter right corresponded to Article 3 ECHR, so “the meaning and scope of the rights are the same”, as set out in Article 52(3) of the Charter.  So the ECJ followed the case law of the ECtHR on Article 3 ECHR, referring specifically to the revised test on “medical cases” set out in Paposhvili, and adding that when applying Article 4 of the Charter, “particular attention must be paid to the specific vulnerabilities of persons whose psychological suffering, which is likely to be exacerbated in the event of their removal, is a consequence of torture or inhuman or degrading treatment in their country of origin”.

It followed that the Charter, interpreted in light of the ECHR, “preclude[s] a Member State from expelling a third country national where such expulsion would, in essence, result in significant and permanent deterioration of that person’s mental health disorders, particularly where, as in the present case, such deterioration would endanger his life.” It also recalled its previous ruling on “medical cases” and the Returns Directive.

But since the courts in the UK had already ruled out MP’s removal, the non-removal point was not relevant. Rather the issue was whether MP is entitled to subsidiary protection. Here the ECJ recalled its prior ruling that “medical cases” were not normally entitled to subsidiary protection, but noted that M’Bodj concerned a victim of assault in the host Member State, whereas MP was tortured in the country of origin and the after-effects would be exacerbated in the event of return. Both of these factors are relevant when interpreting the qualification Directive; but “such substantial aggravation cannot, in itself, be regarded as inhuman or degrading treatment inflicted on that third country national in his country of origin, within the meaning of” the Directive.

What about the lack of medical care for after-effects of torture in the country of origin? The Court reiterated its position that a right to subsidiary protection “cannot simply be the result of general shortcomings in the health system of the country of origin”, and that “deterioration in the health of a third country national who is suffering from a serious illness, as a result of there being no appropriate treatment in his country of origin, is not sufficient, unless that third country national is intentionally deprived of health care, to warrant that person being granted subsidiary protection”.

But on this point, it was crucial that this was not an “ordinary” example of a “medical case”, but one deriving from torture. The preamble to the qualification refers to taking into account international human rights law considering the subsidiary protection definition; and so the ECJ interpreted the UN Convention Against Torture (UNCAT) for the first time in its case law. In particular, the Court examined Article 14 of that Convention, which gives torture victims a right to redress and rehabilitation.

Overall, the Court insisted on a separation between UNCAT and refugee law, by analogy with the distinction between refugee law and the international law of armed conflict (the Geneva Conventions) which it had previously insisted upon in its judgment in Diakité. This was because the UNCAT system and refugee law pursue different purposes. So it followed that:

…it is not possible, without disregarding the distinct areas covered by those two regimes, for a third country national in a situation such as that of MP to be eligible for subsidiary protection as a result of every violation, by his State of origin, of Article 14 of the Convention against Torture.

So not every violation of Article 14 of UNCAT leads to subsidiary protection. But that implies that some violations do. The Court went on to clarify:

It is therefore for the national court to ascertain, in the light of all current and relevant information, in particular reports by international organisations and non-governmental human rights organisations, whether, in the present case, MP is likely, if returned to his country of origin, to face a risk of being intentionally deprived of appropriate care for the physical and mental after-effects resulting from the torture he was subjected to by the authorities of that country. That will be the case, inter alia, if, in circumstances where, as in the main proceedings, a third country national is at risk of committing suicide because of the trauma resulting from the torture he was subjected to by the authorities of his country of origin, it is clear that those authorities, notwithstanding their obligation under Article 14 of the Convention against Torture, are not prepared to provide for his rehabilitation. There will also be such a risk if it is apparent that the authorities of that country have adopted a discriminatory policy as regards access to health care, thus making it more difficult for certain ethnic groups or certain groups of individuals, of which MP forms part, to obtain access to appropriate care for the physical and mental after-effects of the torture perpetrated by those authorities.

So there are two cases where subsidiary protection would apply, due to intentional deprivation of care: the authorities are “not prepared” to fulfil their UNCAT obligations of rehabilitation to a person at risk of suicide following from torture suffered in that country; or there is discriminatory policy “making it more difficult” for certain groups to obtain such treatment. These criteria are non-exhaustive (“inter alia”). The evidence to be considered to this end is “all current and relevant information, in particular reports by international organisations and non-governmental human rights organisations”. Again, the sources of evidence are non-exhaustive (“in particular”).


At first sight, the Court’s judgment sticks to the framework developed in its prior case law: there is no right to subsidiary protection in “medical cases”, except where care is deliberately refused. But look closely, and it’s clear that the Court has developed that case law in important ways in today’s judgment.

First of all, the definition of “medical cases” is now wider, since the Court explicitly adopts the revised interpretation of Article 3 ECHR from recent ECtHR case law. Secondly, in torture cases, the Court has elaborated what factors to consider to determine if inadequate health care would be intentionally withheld in the country of origin. If the asylum seeker is suicidal due to the after-effects of torture carried out in that country, then if that country is either “not prepared” to fulfil UNCAT obligations of rehabilitation to such persons or has a discriminatory policy “making it more difficult” for certain groups to obtain care would amount to an “intentional” deprivation of health care, there is a right to subsidiary protection. The first of these grounds is unique to torture victims, but the second ground should arguably be relevant to any “medical cases”.

Thirdly, the Court has fleshed out the back-up obligation of non-removal for “medical cases” even in the event that subsidiary protection is not granted, insisting that it is an EU law obligation based on the Charter, alongside its prior ruling that the Returns Directive rules it out. This is particularly relevant for the UK and Ireland, given that they are not covered by the Returns Directive. In fact it is not obvious at first sight how EU law – and therefore the Charter – applies in those countries to such cases, if the persons concerned have no right to subsidiary protection. Arguably the link to the grounds for subsidiary protection set out in the qualification Directive is sufficient; but the Court should have spelled this out.

In the Member States bound by the Returns Directive, the finding that the Charter applies to prevent such removal simplifies the process of guaranteeing the non-removal of “medical cases”. Furthermore, it should be recalled that the case law on that Directive guarantees health care and medical assistance.

Overall, then, today’s judgment has gone some way to ensuring greater protection, where necessary, for the most vulnerable migrants: torture victims and the terminally ill.

Barnard & Peers: chapter 9, chapter 26

JHA4: chapters I:5, I:7
Photo credit: The Guardian Nigeria

Thursday, 19 April 2018

Windrush: Violating data protection law under the guise of protecting it

Matthew White, PhD candidate Sheffield Hallam University.


There have been numerous reports of Windrush Generation Commonwealth citizens being denied health care, detained, losing jobs and threats of deportation. Nick Nason describes the Windrush Generation as Commonwealth citizens from the West Indies who were invited to the UK after WWII to address the shortage of workers at the time. There was a time when West Indians enjoyed total freedom of movement. Nason notes that s.2(2)(b) of the Commonwealth Immigrants Act 1962 exempted from immigration controls those who arrived with their parents and were under 16. And this is still true for children who arrived prior to 1 January 1973, as Nason puts it, they are ‘in the UK legally.’   

The issue for the Windrush Generation arises due to successive immigrations laws, the 2014 and 2016 Immigration Acts. Both are designed to create a ‘hostile environment’ to ‘to make life so difficult for individuals without permission to remain that they will not seek to enter the UK to begin with or if already present will leave voluntarily.’ These new Acts required proof of one’s right to be in the UK, and would be denied access to key services (see above) if there was no evidence of this. It is this denial of access to services that has brought this shameful chapter in British history to light to the point where the Prime Minister, Theresa May had to apologise to Caribbean leaders. The sincerity of said apology is open to question given that vital protections for the Windrush Generation were removed from the 2014 Act and were warned about the implications of Act in question.

Destroying personal data on data protection grounds

The Home Office then relied upon data protection law to justify deletion by arguing that keeping personal data for longer than necessary was in breach of data protection principles. More specifically, Robert Peston tweeted that the Home Office relied upon the Fourth and Fifth data protection Principles found in Schedule 1 of the Data Protection Act 1998 (DPA 1998).

The actions of the Home Office in relation to the destruction of personal data does not just have implications with regards to the DPA 1998 but also under the European Convention on Human Rights (ECHR), particularly Article 8 which provides that:

1.      Everyone has the right to respect for his private and family life, his home and his correspondence.

2.      There shall be no interference by a public authority with the exercise of this right except such as in accordance with the law and is necessary in a democratic society in the interests of national security, public safety or the economic well-being of the country, for the prevention of disorder or crime, for the protection of health or morals, or for the protection of the rights and freedoms of others.

According to the European Court of Human Rights (ECtHR) Grand Chamber (GC) in S and Marper (ECHR, 4 December 2008) the ‘protection of personal data is of fundamental importance to a person’s enjoyment of his or her right to respect for private and family life, as guaranteed by Article 8 of the Convention’ [103]. The mere storage of personal data interferes with Article 8 [67]. The GC continued that ‘domestic law must afford appropriate safeguards to prevent any such use of personal data as may be inconsistent’ with Article 8 [103]. The Home Office’s position on not storing personal data for longer than is necessary is consistent with the ECtHR’s approach [ibid], but this would be classed as subsequent [67, 121] use and thus is still an Article 8 issue.

The first requirement under Article 8 is whether the deletion of personal data was ‘in accordance with the law.’ This requires there to be some basis in domestic law [193]. One could argue the DPA 1998 itself provides the domestic law basis for deletion, but the ECtHR has previously held that it does not have to assess ‘the quality of the applicable data protection framework in the abstract and must rather confine itself as far as possible to examining the particular consequences of application of its provisions in the case before it’ [81]. This is due to the fact that reliance on the DPA 1998 does not guarantee an action to be ‘in accordance with the law’ [207]. The ECtHR has stressed that applicable laws must provide:

[C]lear, detailed rules governing the scope and application of the relevant measures; as well as minimum safeguards concerning, inter alia, duration, storage, usage, access of third parties, procedures for preserving the integrity and confidentiality of data and procedures for their destruction, thus providing sufficient guarantees against the risk of abuse and arbitrariness at each stage of its processing [75].

Therefore, the legal basis for the destruction of personal data in the context to which the Home Office relies becomes severely weakened. The DPA 1998 does not define the scope and application with clear, detailed rules as to when the Home Office is entitled to delete personal data, nor does it provide procedures for said destruction. The arbitrariness of the measure is apparent when it is clear that the Home Office deleted all said personal data en masse.

Moreover, even if one were to consider the DPA 1998 as the correct legal basis that is sufficient in ECHR terms, this does not answer the question as to why the Fourth Principle was used in this manner. The Home Office are essentially arguing that personal data held on Windrush Generation individuals were inaccurate, without actually taking reasonable steps to ensure the accuracy of said data in contravention of Schedule 1, Part II (7)(a) of the DPA 1998. When the domestic authorities do not even observe their own law, this would also violate Article 8 [45-9]

The lawful basis in this context is strongly linked to whether a measure satisfies the ‘quality of the law’ in which a law should be accessible to the person concerned and foreseeable to its effects [50]. This is usually satisfied when a law is published [52-3]. However it has been argued that the vagueness of the DPA 1998 provides an insufficient legal basis for the destruction of personal data in this context. In arguing so, it cannot be said the law is accessible, because there is no law to access, which in and of itself would violate Article 8 [69-70].

Regarding foreseeability, this is described as formulating the law:

[W]ith sufficient precision to enable the individual – if need be with appropriate advice – to regulate his conduct. For domestic law to meet these requirements, it must afford adequate legal protection against arbitrariness and accordingly indicate with sufficient clarity the scope of discretion conferred on the competent authorities and the manner of its exercise [95].

The level of precision ‘depends to a considerable degree on the content of the instrument in question, the field it is designed to cover and the number and status of those to whom it is addressed’ [96]. The DPA 1998 was designed to cover the protection of (sensitive) personal data, but not specifically in the immigration context, thus its Principles are not precise [98]. The DPA 1998 would not indicate to any Windrush Generation individual as to when or under what circumstances their personal data may be deleted by the Home Office, thus not providing sufficient clarify on the scope of their discretion.

Again, the arbitrariness of the Home Office’s actions is apparent when it destroyed thousands of landing card slips in 2010. For example, when would it be necessary to delete landing card slips? Would it be when the Home Office could guarantee that an individual would no longer require it to demonstrate they came as a child before 1973? It would be contrary to the rule of law if the Home Office used its power in an unfettered manner [62]. The exercise of power by the Home Office ‘was arbitrary and was based on legal provisions which allowed an unfettered discretion to the executive and did not meet the required standards of clarity and foreseeability’ thus amounting to a violation [86, 89].

This discussion on the unlawfulness of the Home Office’s reliance could have stopped at the end of the last paragraph, but it is important to consider the case of Kurić and others v Slovenia (ECHR, 13 July 2010) as it shares similarities with the Windrush Generation scandal. The applicants in this case complained before the ECtHR that the erasure of their names from the Register of Permanent Residents made them aliens overnight which denied them ‘civil, political, social and economic rights’ [319].

The applicants had been living in Slovenia for years, and most of them decades, some were even born there [356]. The applicants did not enter Slovenia as immigrants but as settled citizens [357]. Moreover, the applicants had a stronger residence status than long-term migrants and those seeking to enter or remain [357]. Although not identical, the erasure of landing cards made it more difficult for Windrush Generation individuals to prove they had a right to live in the UK, and due to this lack of proof they could be denied healthcare, jobs, bank accounts etc.

In that case, the ECtHR reiterated previous case law in that Article 8 is interfered with when the ‘persons concerned possess strong personal or family ties in the host country which are liable to be seriously affected by application of the measure in question’ [351]. They continued that the right to establish and develop relationships, embracing social identity, having social ties with the community all fall within the meaning of Article 8 [352]. Moreover, Article 8 is interfered with when one faces expulsion and having their citizenship arbitrarily denied [352-3]. Finally, the UK Government has positive obligations to respect Article 8 [354].

Due to the cumulative failings of Slovenia, the ECtHR concluded there was a violation of Article 8 [376]. The ECtHR did not decide whether the Article 8 violation was due to the measures not being ‘in accordance with the law’ pursued a legitimate aim or was ‘necessary in a democratic society,’ so the same approach will be taken to argue that in the cumulative, Article 8 has been violated. When one considers that landing cards had been destroyed arbitrarily in 2010, the Home Office claimed these had no impact on the rights of the Windrush Generation. This claim is contested by two Home Office whistle blowers arguing that the landing cards had been a useful resource. The whistle blower’s account is supported by the Border Force where its notes state that ‘Information from a landing card may be used by an entry clearance officer in making a decision on a visa application.’ Destroying landing cards allowed Home Office staff to tell those concerned that they had no record of arrival dates which would lead to the denial of services and at worst, deportation.

Moreover, citing data protection law as a reason for the destruction of personal data appears cynical due to the amount of personal data that is kept anyway and the fact that the same Government is seeking to create an immigration exemption in the new Data Protection Bill (Schedule 2, Part 1, (4). The Home Office also explained that it considers alternative evidence such as tax records, utility bills and tenancy agreements as evidence of ongoing residency. However, if one can be denied work, have bank accounts frozen and be denied tenancy, then this evidence could also be difficult to provide. The cumulative effect of denial of services to the threat of (or actual) deportation, the deletion of flying cards and the spurious reasoning behind it would amount to a violation of Article 8.

The racist elephant in the room

Nason asked whether the overt racism from the 1960s-80s has simply been replaced ‘by a more insidious, state-endorsed hostility in the name of immigration control.’ A group of NGOs published a report on the ‘hostile environment’ noting that its very nature is discriminatory and thus encourages discriminatory or even racist behaviour. Former Home Office employees detail how the ‘hostile environment’ changed the attitude of staff to the point where they enjoyed catching out Windrush individuals without evidence. James Moore argues that this is what happens when you let dog-whistle racism go mainstream.

Article 14 of the ECHR details how the enjoyment of rights contained in the ECHR must be protected in a non-discriminatory manner. The grounds for discrimination are non-exhaustive but include race, colour, national or social origin and birth. Any one of these can be relevant to the Windrush Generation. Article 14 only works in combination with another substantive Convention Right, in this instance Article 8 [84]. Article 14 requires a difference in treatment to those in an analogous or similar situation [66]. The ECtHR have maintained that:

[A] difference in treatment may take the form of disproportionately prejudicial effects of a general policy or measure which, though couched in neutral terms, discriminates against a group…may amount to “indirect discrimination”, which does not necessarily require a discriminatory intent [184].

The Windrush Generation have as much right to be here as any other UK citizen, yet they are the ones that a targeted under the ‘hostile environment.’ Given that the Home Office destroyed landing cards, removed key protections that could have avoided this. One could argue the actions of the UK Government are more than just indirect discrimination because the discriminatory intent arises for the poor reasoning for destruction of flying cards to the lack of reasoning for removing key protections. The Government has no objective reasonable justification for this difference in treatment and thus amounts to discrimination [196]. Moreover, the Windrush Generation are being treated as though they are immigrants. This engages a different type of discrimination issue under Article 14, a Thlimennos discrimination which notes that:

The right not to be discriminated against in the enjoyment of the rights guaranteed under the Convention is also violated when States without an objective and reasonable justification fail to treat differently persons whose situations are significantly different [44].

There is no objective reasonable justification on any of the discriminatory grounds and thus amounts to a violation of Article 14 in conjunction with Article 8 [208-210]. Thus, under the ECHR, the racist and discriminatory elephant in the room is glared upon with distain.


This post has highlighted that the dubious reasoning as to why the Home Office destroyed crucial information that could have helped prevent some of the tragedies of the Windrush Generation is flawed, logically and legally. Not only is it flawed, reliance on data protection grounds in ECHR terms would amount to using Article 8 as a shield, and for the UK Government to do so would raise serious questions regarding Article 17 (the abuse of rights). Not only is the Home Office’s actions and reasoning in violation of Article 8, it violates Article 8 on the ground of defending it. There is a bigger issue which highlights the resurfacing of racism and discrimination in a new form which violates Article 8 in conjunction with Article 14. The ‘hostile environment’ has provided a platform and has legalised discrimination and racism, the destruction of landing cards in 2010 can be seen as the first steps towards this, and the removal of key protections for the Windrush Generation in the 2014 Act  is no accident either. The ‘hostile environment’ is the problem and the recent outrages shows that there are ‘resources of hope, but time is running out – we are at five minutes to midnight.’

Photo credit: www.sas.ac.uk

Wednesday, 18 April 2018

Salvation outside the church? The ECJ rules on religious discrimination in employment

Dr Ronan McCrea, Senior Lecturer in Law, University College London


The Court of Justice has issued its first major ruling on the reconciliation of the autonomy rights of religious organisations with the right of employees (or potential employees) of such organisations to be free of discrimination.

In 2012 Vera Egenberger applied for a fixed term post advertised by the Evangelisches Werk für Diakonie und Entwicklung, which is a body associated with the Evangelische Kirche in Deutschland (a German Protestant church). The post advertised sought a person who could prepare a report on Germany’s compliance with the United Nations International Convention on the Elimination of All Forms of Racial Discrimination. Ms. Egenberger had significant experience in this area and applied for the post. However, there was a problem. Ms. Egenberger is a person who does not have a religious faith and the relevant advert included the following statement:

‘We require membership of a Protestant church, or of a church which is a member of the Arbeitsgemeinschaft Christlicher Kirchen in Deutschland (Cooperative of Christian Churches in Germany), and identification with the welfare mission. Please state your membership in your curriculum vitae.’

Ms. Egenberger was not called for interview. She took a case in the German courts alleging discrimination on grounds of religion.

As discrimination in employment on grounds of religion is regulated by EU law, in the form of Directive 2000/78 (which also bans discrimination on grounds of disability, age or sexual orientation in employment), when the case came before the Bundesarbeitsgericht (Federal Labour Court) it decided to make a reference to the Court of Justice to clarify the interpretation of EU law. The key issue in the reference was whether the scope of the exemption from the duty not to discriminate on grounds of religion or belief granted by German law to religious organisations was compatible with Directive 2000/78.

The issue of exemptions from the prohibition on discrimination on grounds of religion for religious employers is addressed by Article 4(2) of the Directive which states:

‘…. in the case of occupational activities within churches and other public or private organisations the ethos of which is based on religion or belief, a difference of treatment based on a person's religion or belief shall not constitute discrimination where, by reason of the nature of these activities or of the context in which they are carried out, a person's religion or belief constitute a genuine, legitimate and justified occupational requirement, having regard to the organisation's ethos. This difference of treatment shall be implemented taking account of Member States' constitutional provisions and principles, as well as the general principles of Community law, and should not justify discrimination on another ground.’ (emphasis added).

The relevant German law implementing the directive provided that:

‘….a difference in treatment based on religion or belief shall also be admitted in the case of employment by religious societies, by institutions affiliated therewith, regardless of legal form, or by associations whose purpose is to foster a religion or belief in the community, where a given religion or belief constitutes a justified occupational requirement, having regard to the employer’s own perception, in view of the employer’s right of autonomy or by reason of the nature of its activities.’ (paragraph 9(1) of the Allgemeine Gleichbehandlungsgesetz, emphasis added).

This legislation, has been interpreted in the light of the German constitutional guarantee that states:

‘Religious societies shall regulate and administer their affairs independently within the limits of the law that applies to all. They shall confer their offices without the involvement of central government or local authorities.’ (Grundgesetz Article 140).

This has meant that the consistent approach of the German courts has been that the decision as to whether a particular role within a religious organization needs to be limited to those of a particular faith was for the religious employer to take. The role of the courts has been limited to plausibility review, on the basis of a religion’s self-conception defined by belief.

The national court harboured doubts as to whether the approach of German law in allowing the religious employer to determine for itself, subject only to plausibility review by the courts, whether its beliefs required a particular role to be reserved to those of a particular faith, was compatible with the directive and therefore made a reference to the Court of Justice under Article 267.

The Ruling: A More Balanced Approach Needed

The Court of Justice’s ruling made it clear that German law had gone too far by allowing such a wide scope for religious employers to determine for themselves whether a particular job could be reserved to those of a particular faith.

It noted that Article 4(2) of the Directive allowed the discrimination on grounds of religion only if having regard to the nature of the activity concerned or the context in which it is carried out, ‘religion or belief constitute[s] a genuine, legitimate and justified occupational requirement, having regard to the organisation’s ethos’ and concluded that:

‘if review of compliance with those criteria were, in the event of doubt as to that compliance, the task not of an independent authority such as a national court but of the church or organisation intending to practise a difference of treatment on grounds of religion or belief, [this provision of the Directive] would be deprived of effect.’

Interestingly, although the employer had cited both the guarantee of freedom of religion or belief (Article 10 of the Charter of Fundamental Rights) and Article 17 of the Treaty on the Functioning of the European Union, which provides that the Union ‘The Union respects and does not prejudice the status under national law of churches and religious associations or communities in the Member States’, the Court also relied heavily on constitutional principles to bolster its conclusion that excessive leeway had been granted to religious employers by German law.

The Court noted that Directive 2000/78 was merely a ‘specific expression, in the field covered by it, of the general prohibition of discrimination laid down in Article 21 of the Charter’ (which sets out a general ban on discrimination). It also noted that that when an individual establishes before a national court facts from which it may be presumed that there has been direct or indirect discrimination then, under Article 10 of the Directive, it is for the respondent to prove that there has been no breach of that principle. Thus, the need under Article 47 of the Charter to provide effective judicial protection of EU law rights meant that restricting the ability of the national courts to review the decision of an employer to impose a discriminatory requirement would be contrary to EU law.

Next, the Court held that the objective of Article 4(2) of the Directive was to ensure “a fair balance” between the autonomy rights of religious organisations and the right of workers to be free from discrimination. The Directive “sets out the criteria to be taken into account in the balancing exercise” and in the event of a dispute it must be possible for the balancing exercise to be reviewed by a national court. For the Court, the commitment to respecting the status of religious organisations in Article 17 of the Treaty could not change this conclusion.

That article’s function was:

‘to express the neutrality of the European Union towards the organization by the Member States of their relations with churches and religious associations and communities […] [and] is not such as to exempt compliance with the criteria set out in Article 4(2) of Directive 2000/78 from effective judicial review.’

Guidance on the Test to Be Applied

Having found that the German legislation was not compatible with the Directive the Court then had to address two further interesting issues. First, it had to give guidance to the national court on the question of how the ‘fair balancing’ ought to be carried out and then it had to advise on how to implement the consequences of its finding in the case.

In relation to how to carry out the ‘fair balancing’ the Court had to walk a tightrope. It acknowledged that under ECHR case law, states are precluded from assessing the legitimacy of the beliefs of a religious organization. However, it also had to ensure that the imposition of an occupational requirement relating to religion or belief was, in the words of the Directive ‘genuine, legitimate and justified, having regard to [the] ethos [of the religious employer]’. Thus it had to decide how to recognize the necessarily subjective requirements of the ethos of the employer, with the seemingly objective requirements of the ‘genuine, legitimate and justified’ test.

The Court adopted an approach that is significantly more objective than the approach taken in German law. It set out a test under which religious organisations must show an ’objectively verifiable existence of a direct link between the occupational requirement imposed by the employer and the activity concerned.’ Thus, in order to meet Article 4(2)’s requirements that the difference in treatment on grounds of religion be ‘genuine, legitimate and justified’ the Court held that:

‘To be considered ‘genuine’: ‘professing the religion or belief on which the ethos of the church or organisation is founded must appear necessary because of the importance of the occupational activity in question for the manifestation of that ethos or the exercise by the church or organisation of its right of autonomy.’

To be considered ‘legitimate’ it found that the national court must ‘ensure that the requirement of professing the religion or belief on which the ethos of the church or organisation is founded is not used to pursue an aim that has no connection with that ethos or with the exercise by the church or organisation of its right of autonomy.’

And to be considered justified the CJEU set down that ‘the church or organisation imposing the requirement is obliged to show, in the light of the factual circumstances of the case, that the supposed risk of causing harm to its ethos or to its right of autonomy is probable and substantial, so that imposing such a requirement is indeed necessary.’

Finally, although a proportionality requirement is not included in the text of Article 4(2) (and is included in other Articles of the Directive), the Court held that as proportionality is a general principle of EU law, the exemption given by Article 4(2) is to be read as being subject to a proportionality requirement.

Applying the Ruling

Given the possibility of a clash between German law and the requirements of the Directive the German court asked for guidance on how it should proceed if it proved impossible to interpret domestic law so as to comply with the Directive (bearing in mind the contra legem exception in the Marleasing line of case-law on the indirect effect of Directives; ie a national court cannot be required to interpret national law consistently with a Directive to the extent of ignoring the express wording of national law).

The Court seemed to doubt that an interpretation consistent with EU law was impossible, noting that the duty to interpret national law consistently with EU law included a duty for national courts ‘to change their established case-law where necessary’ (referring to the DI judgment on age discrimination, discussed here). However, it went on to say that should consistent interpretation prove impossible then the Court should disapply national law and give effect to the relevant EU law rights itself.

It justified this position on the basis that Directive 2000/78 did not establish the right to equal treatment. Rather it sets out a framework for combatting discrimination on various grounds. The right to equal treatment is, the Court held, a general principle of law and is enshrined in Article 21 of the Charter. Given that Article 47 of the Charter requires that adequate judicial protection be given to such rights, national courts have to ensure ‘the judicial protection deriving for individuals from Articles 21 and 47 of the Charter and to guarantee the full effectiveness of those articles by disapplying if need be any contrary provision of national law.’ This develops earlier case law on the issue of when the Charter itself does (and does not) have direct effect, in particular the AMS case discussed here; and it confirms the UK case law on the direct effect of Article 47 (Vidal-Hall and Benkharbouche, discussed here and here).


The Court of Justice has given a notably constitutionalized interpretation of the Directive in this case. It has relied on the Charter and general principles of law to read a proportionality test into Article 4(2) that did not appear in the text. It has insisted on proportionality as the framework within which the ‘fair balancing’ of the autonomy rights of religious employers and equal treatment rights of employees must take place.

This approach is in tension with recent trends in the caselaw of the Strasbourg Court. In cases such as Fernandez Martinez v Spain, the Court of Human Rights had moved away somewhat from the balancing of rights seen in its earlier caselaw and had moved closer to the ‘ministerial exemption’ model used in the United States, under which religious organisations have an absolute exemption from non-discrimination laws in respect of roles that include religious functions. Given the strong emphasis placed on proportionality by the EU legal order, unsurprisingly, in Egenberger, the Court of Justice has clearly come down in favour of the balancing approach (see R. McCrea “Singing from the Same Hymn Sheet? What the Differences between the Strasbourg and Luxembourg Courts Tell Us about Religious Freedom, Non-Discrimination and the Secular State” Oxford Journal of Law and Religion (2016) 5(2) 183-210, 198-99).

The Court’s insistence that EU non-discrimination law merely codifies a self-executing (and horizontally directly effective) constitutional general principle of non-discrimination law represents a continuation of the controversial line of cases beginning in Mangold which has attracted significant criticism, including from national courts given the potential for legal uncertainty that such an approach involves. Here, the Court of Justice has made it clear that the EU’s constitutional commitment to proportionality means that religious bodies may only impose discriminatory conditions on employees when it is proportionate to do so and national courts must be empowered to ensure religious employers do not exercise their right to discriminate in a disproportionate way.

This is in tension with the approach adopted by the German legislature which, in the light of German constitutional guarantees of religious autonomy, gave very restricted powers to the courts to second guess the decisions of religious bodies in this way. Given that EU and German constitutional norms appear to be in tension with each other in this way it will be interesting to see how this ruling is applied by the national court. The German constitutional provisions on religious autonomy go all the way back to the Weimar constitution and are taken very seriously, though it may be an exaggeration to view them as constituting the kind of core ‘constitutional identity’ that might trigger a refusal by the German courts to give primacy to EU law.

Barnard & Peers: chapter 9, chapter 20

Photo credit: Wikimedia Commons

Friday, 13 April 2018

Childhood’s End? The Court of Justice upholds unaccompanied child refugees’ right to family reunion

Professor Steve Peers, University of Essex

Turning 18 is a big moment in any young person’s life. Although it rarely entails, by itself, an immediate change in their social and economic links with their parents, it is widely recognised as a significant rite of legal passage, marking as it does the official date of becoming an adult.

But what if the main legal impact of turning 18 is not the enhancement of a young person’s legal rights, but rather their deterioration? That is often the scenario in immigration or asylum law, in particular for those who need protection the most: unaccompanied minors. Since immigration and asylum procedures often take some time, the question then arises what happens if applicants are underage when a process began, but become an adult before it finishes. Do they retain throughout that process the special legal protection accorded to children? At what point exactly does that special legal status end?

That was the issue in yesterday’s judgment in A and S, which was the first time the Court of Justice has ruled on the family reunion rights of child refugees. The judgment concerns the EU’s family reunion Directive, which contains special rules for the family reunion of refugees in general, and unaccompanied minor refugees in particular.  However, it is possible that it has an impact on the status of young people in EU immigration and asylum law more generally.

The basic EU rules on family reunion

The EU’s family reunion Directive sets minimum standards, so states can be more generous if they wish. It mainly concerns reunion of spouses and minor children with a non-EU sponsor; admission of further family members is optional in most cases. It does not apply to the UK, Ireland and Denmark. However, it will apply to family reunion of UK citizens in the EU (besides those living in Ireland and Denmark) after the post-Brexit transition period, when the UK is no longer covered by EU free movement law, unless (a) they are covered by the withdrawal agreement (see discussion here), or (b) the EU (or, if legally possible, individual Member States) and the UK agree special rules on post-Brexit family migration.

The standard rules in the Directive require that: the sponsor has a residence permit valid for at least one year, and has “reasonable prospects” of obtaining permanent residence; the family members must reside outside the territory when the application is made (although Member States can derogate from that rule); “public policy, public security or public health” are grounds for rejection; conditions relating to accommodation, sickness insurance and “stable and regular resources” may be imposed; Member States may require “integration measures”; and there can be a waiting period of two years of lawful stay of the sponsor before family reunion takes place.

There are also exclusions from the scope of the Directive. It does not apply at all to: asylum seekers; persons with temporary protection; persons with subsidiary protection on the basis of national or international law; and family members of EU citizens (whether they have moved within the EU or not). Implicitly it does not apply to irregular migrants, since by definition they do not have a residence permit with the prospect of long-term residence, until and unless Member States decide to regularise their status.   

Member States can set lower standards than the Directive, where it allows for such derogations, although this is subject to detailed conditions. These derogations exist as regards: children over 12, who arrive separately from the rest of the family; minimum ages for the sponsor or spouse; children over 15; and a waiting period of three years.

The Court of Justice has ruled on the Directive several times, as regards: its validity in light of human rights concerns (EP v Council); its application to dual EU/non-EU citizens (O and S) the sufficient resources condition (Chakroun and Khachab); the minimum age of spouses (Noorzia, discussed here); and the integration conditions (K and A, discussed here). Pending cases concern: the application of the Directive by analogy to family reunion with “home State” EU citizens (C and A) and persons with subsidiary protection (K and B and E); the rejection of a separate residence permit due to failure to comply with integration conditions (K); and loss of a residence permit due to fraud which the family member was unaware of (YZ and others).

As well as the special rules for refugee family reunion set out in the original Directive, subsequent EU legislation contains more favourable rules for the family reunion of other groups of non-EU citizens: holders of an EU Blue Card for highly-skilled workers (discussed here); intra-corporate transferees (discussed here); and researchers (discussed here). The proposal to amend the Blue Card law (discussed here) would enhance these rules further. Yesterday’s judgment is the first time the ECJ has interpreted any of these special rules.

Exceptions for refugees

The special rules apply to a refugee who has been “recognised” by a Member State, meaning that their asylum application for refugee status in that State was successful. A “refugee” is defined not by reference to EU law, but to international law – the UN (Geneva) Convention on the Status of Refugees and its protocol – since the Directive was adopted before the EU adopted its own asylum laws. Member States. Member States may limit the special rules to family relationships which predate entry to the Member State.

These rules waive a number of conditions for family reunion: the optional derogation for those over 12; the conditions relating to accommodation, sickness insurance and “stable and regular resources” (although Member States can apply those conditions if the sponsor or family members have “special links” with a non-EU country, or if the application was submitted more than three months after refugee status was granted); and the waiting period. The rules on evidence of family relationships are also relaxed, in the event that documentary evidence is unavailable. Finally, the optional “integration measures” requirement can only be applied after family reunion, whether the family relationship existed before entry or not. 

Conversely, other conditions still apply: the requirement of a residence permit valid for at least one year, with “reasonable prospects” of obtaining permanent residence; residence outside the territory when the application is made; and “public policy, public security or public health”. Satisfying the first of these criteria is made easier by EU law, since the Qualification Directive requires refugees, once their status is recognised, to receive a residence permit valid for at least three years, and refugees can qualify for EU long-term residence status under the relevant Directive.

Most importantly for our case, the refugee rules make the admission of the parents of unaccompanied minor refugees mandatory, rather than optional; and they waive the conditions that otherwise apply to the admission of migrants’ parents (they must be “dependent” on the sponsor and they “do not enjoy proper family support in the country of origin”).

The judgment

The A and S case concerns a young Eritrean woman who arrived in the Netherlands and made an asylum application just before turning 18. Her application was successful after her birthday, and a NGO shortly afterward applied on her behalf for admission of her parents and siblings on the basis of the special rules in the family reunion Directive.  But could she rely on the special rules at all – given that she was over 18 when the application for family reunion was made, and indeed when her refugee status was recognised?

The Dutch government argued that the relevant date when a person must be considered a minor should be determined by national law, while the Commission argued for the date of the application for family reunion, and the Polish government argued for the date of the decision on the family reunion application. The applicants (the young woman’s parents) argued for the date of her initial entry onto the territory. No one argued for another reasonable possibility: the date of the decision on the refugee application (although that would raise the question of what the date would be if that decision was appealed). Ultimately the Court decided that the relevant date was the date of applying for refugee status.

The Court’s starting point was (as it had ruled before) the “right” of family reunion guaranteed by the Directive, which the addition of the intention (in the preamble of the Directive) to ensure “more favourable conditions for refugees for the exercise of” that right, “on account of the reasons which obliged them to flee their country and prevent them from leading a normal family life there”. Those more favourable rules include a mandatory admission of the parents of unaccompanied minors, waiving the normal conditions which would usually apply.

Next, the Court noted that the definition of “unaccompanied minor” in the Directive was not absolutely fixed at entry: parents could arrive after the child’s entry, or desert the child after entry. In that context, it was unclear from the text of the Directive when the requirement of being 18 had to apply. But that did not mean Member States had discretion to decide that issue; the Court applied the normal rule that in the absence of an express reference to the laws of the Member States, a provision of EU law “must normally be given an autonomous and uniform interpretation throughout the European Union, and that interpretation must take into account, inter alia, the context of the provision and the objective pursued by the legislation in question”.

Since other provisions of the Directive refer explicitly to national law, the absence of such a reference in the definition of “unaccompanied minor” had an a contrario effect. The objective of the Directive was to give a right of admission to their parents, in the context of protecting family life with more favourable conditions for refugees. This case had to be distinguished from Noorzia, on the minimum age of spouses for family reunion, which concerned an optional rule that expressly gave Member States discretion to decide on the age.  Ultimately, then, the issue could not be left to each Member State to determine.

Rather, the uniform definition of “unaccompanied minor” had to be determined “by reference to the wording, general scheme and objective of that directive, taking into account the regulatory context in which it is found and the general principles of EU law”.  As noted already, the wording didn’t settle the issue. The general scheme included the exclusion of asylum-seekers from the scope of the Directive, and the application of the special rules only after the refugee had been “recognised as such by the Member States”. In that context, the Court noted that the EU’s Qualification Directive requires refugee status to be granted if an applicant satisfies the relevant conditions, and states that “recognition of refugee status is a declaratory act”, so that a person who meets the conditions for refugee status “has a subjective right to be recognised as having refugee status…even before the formal decision is adopted in that regard”.

So it followed that the date for assessing the applicant’s age could not be when the decision on refugee status was taken.  Such an interpretation would make status as a minor dependent on the functioning of national administrations, and thus undermine the effectiveness of the family reunion rules and the aims of the Directive, along with “the principles of equal treatment and legal certainty”. That’s because two different children of the same age who applied for asylum would be in a different position depending on how quickly their application was processed, an issue which was outside their control – governed rather by Member States’ decisions about organising their administration. In any event, due to “substantial surges” in asylum applications, decision making might be long winded and “time limits laid down in that regard by EU law are often exceeded”, so a “substantial proportion of refugees” who are unaccompanied minors might be denied their family reunion right. (Note that, with respect, the Court is confused here: the rules in the EU’s asylum procedures Directive on time limits to decide on asylum applications don’t apply until July 2018).

Rather than taking up the option in EU law to fast-track such cases, there might be the “opposite effect”, which would frustrate the objectives of EU legislation and the EU Charter rules on rights of the child. Here the Court obliquely recognises the possible cynicism of national interior ministries, which might simply delay deciding on applications until a child turns 18 if that would lead to preventing the admission of parents. Furthermore the Court rules that using the date of the decision on refugee recognition would “undermine legal certainty” for the young applicants as regards their family reunion.

The Court’s preferred interpretation – using the date on which the asylum application was submitted – “enables identical treatment and foreseeability to be guaranteed for all applicants who are in the same situation chronologically”, as the outcome would depend on facts intrinsic to them, not to the efficiency of national administrations.  However, the Court did accept the argument of the Dutch government and the Commission that some time limit should apply. In the judges’ view, a “reasonable time” would “in principle” be three months after the decision on refugee status, matching the optional three-month deadline explicitly set out in the Directive for refugees to make an application for family reunion before the conditions of accommodation, sickness insurance and sufficient resources apply.

Finally, the Court rejected other possible dates to determine the young person’s age: the date of entry into the territory of a Member State had to be rejected because of the link of the family reunion right with refugee status, which could only be granted after an asylum application; and the date of applying for family reunion, or the date of the decision on that application, would infringe the basic logic of the Court’s reasoning.


The core motivation of the Court’s judgment – to give broad effect to family reunion rights in general, and to the special family reunion rights of child refugees in particular – reflects a rights-based reasoning, rather than the control-based approach taken by many Member States and the EU institutions during the perceived “refugee crisis” of the last few years. Although the Court ties its interpretation of the family reunion Directive closely to the asylum process – even though there was no EU asylum law when the Directive was adopted – it nevertheless views that process with suspicion, as a potential mechanism for frustrating the applicant’s rights. The Court may have an opportunity to develop this line of reasoning further soon, if it is willing to answer questions in the pending cases (referred to above) where the Member State concerned has extended the special rules for refugees in the family reunion Directive to apply also to sponsors with subsidiary protection status (an alternative form of “international protection” which applies where applicants don’t satisfy the criteria for refugee status).

If the Court had fully followed its own logic on the declaratory effect of granting refugee status, then its final conclusion of using the date when the minor applies for asylum is suspect, for the child concerned must have been a refugee either as soon as they entered the territory, or at some later point (likely before they actually applied for asylum) when the situation in their country of nationality or (if stateless) habitual residence changed for the worse. While the Court is right to say that the Directive links the special family reunion rights with refugee status, that link is built in to the Directive anyway because the special family reunion right can never be triggered in the first place unless a successful asylum application is made. In any event, the Court’s judgment means that it is wise for an unaccompanied child who is nearly 18 to apply for asylum as soon as possible after entering the territory, to avoid any risk that they will not be able to invoke the special family reunion rights in the event that their application for refugee status is successful.

What constraints do Member States still retain on family reunion for refugees just turning 18? They can still try to limit access to their territory for the would-be young refugees. However, if those potential refugees make it to the territory, Member States can’t simply ban minors from applying for refugee status in the first place, since the asylum procedures Directive requires that they must be able to apply for asylum one way or another. If refugee status is granted, Member States can use any of the applicable options to restrict family reunion in general or the special refugee rules in particular that they have not already invoked. (Note that some of those options are off the table, since they are subject to a “standstill” rule and so had to be invoked already if they were going to be validly applied).

The Court even gives Member States a new limitation: a possible three month deadline “in principle” for the young refugees to trigger the special rules for their parents to join them. But if the sponsors are subject anyway to the three month deadline to avoid the conditions of accommodation, sickness insurance and sufficient resources, they will need to move quickly in any event. Although refugees have rights to employment in the qualification Directive, it might be hard for a young refugee to find a good enough job in the time available (access to employment for asylum seekers is limited, by the EU’s reception condition directive); and unlike EU free movement law, the family reunion Directive, as confirmed by the case law (see Khachab) requires the sufficient resources to come from the sponsors, not from their family members. In light of the principle of effectiveness, the three month deadline should not apply to those who were wrongly deprived of their family reunion rights before the Court’s judgment (see the recent King judgment on working time holiday pay, by analogy),

Are there broader implications of the judgment for other EU asylum legislation? (There are also special rules on unaccompanied minors in the EU’s returns Directive, concerning irregular migrants). This would be relevant to the qualification Directive, which includes, among other things, an obligation to trace unaccompanied child refugees’ family members. For its part, the asylum procedures Directive grants unaccompanied minors, among other things, exemptions from some procedural limitations; it also sets out rules on the sometimes controversial issue of assessing the age of children in the event of a dispute. The reception conditions Directive also requires some special treatment of unaccompanied minors, including in the context of detention. Finally, the Dublin rules on responsibility for asylum seekers contain a special rule for responsibility for the asylum applications of unaccompanied minors, which the Court of Justice previously interpreted generously.  Like the family reunion Directive, all this legislation has essentially the same definition of “unaccompanied minors” as the family reunion Directive, without addressing the “passage into adulthood” point, so logically ought to be interpreted the same way. (Note that conversely, EU criminal law legislation on child suspects’ rights – discussed here – does explicitly address this issue, setting out rules on this point similar to the Court’s family reunion judgment in its Article 2(3)).

The Court’s judgment might cause political difficulty for Member States, given that the special rules on unaccompanied minors were highly contested when EU refugee legislation was last renegotiated, ending in 2013. (A modest proposal on unaccompanied minors and the Dublin rules, dating from 2014 and discussed here, got nowhere). The issue may well arise again now that the legislation is being revised a further time, in particular as regards the Dublin rules, where (as discussed here), the Commission seeks to overturn the Court’s prior ruling in favour of unaccompanied minors.

Legalese aside, what is the impact of the new judgment for the families of young refugees? It means they can come straight to the Member State where their child lives, without having to go through the Dublin process first. (Although the Dublin rules allocate responsibility to the State where a family member is a refugee, there are sometimes problems applying this in practice; and the Commission proposal to amend the Dublin rules seeks to undercut those family rules indirectly). When they arrive, they will have access in principle to rights of access to employment et al on the same basis as their young refugee family member, as set out in the qualification Directive.

But most fundamentally, the ruling means that family members have safe passage: the obligation to give them authorisation for legal entry means they can travel to the EU without having to pay smugglers and risk mistreatment or drowning en route. So it’s no exaggeration to say that this judgment could literally save the lives of the parents of some vulnerable young refugees.

Barnard & Peers: chapter 26

JHA law: chapter I:5, I:6

Photo credit: care4calais.org