Karin de Vries (professor of fundamental rights law, Utrecht University)
and Sarah Ganty (JSD candidate, Yale Law School;
FNRS Post doc fellow, UCLouvain)
Photo credit: Kristoffer Trolle, via Wikimedia
Commons
The so-called Danish ‘ghetto law’
case has been one of the most closely watched judgments of 2025 before the
Court of Justice of the European Union (CJEU), long awaited by lawyers working
in EU anti-discrimination law and beyond. Its resonance has extended well
beyond legal circles, attracting sustained attention from the media
and NGOs.
Much has already been written on
the judgment, from different perspectives (for example, here, here,
here
and here).
The focus of this contribution is on how the judgment shapes the definition of racial
discrimination in EU law, including the emerging tension between objective and
social understandings of ethnic origin, the Court’s application of the concepts
of direct and indirect discrimination and its reticence to address the
structural racism at the roots of the Danish housing policy. Before proceeding
to our analysis we first offer a brief outline of the case and the Court’s
reasoning.
Background:
the Danish ‘ghetto law’
At the heart of the case lies a
practice of state-driven gentrification as part of the Danish housing policy. The
Danish Law on public housing seeks to reduce the proportion of
public housing in certain designated neighbourhoods, officially labelled
“transformation areas”. Under the scheme, authorities may sell buildings to
private developers, demolish existing housing, or convert family dwellings into
accommodation for young people. These measures may entail the unilateral
termination of tenants’ leases, resulting in their forced displacement.
To determine
which neighbourhoods are subject to restructuring, an earlier version of the
Law on public housing introduced the labels of “ghettos” and “hard ghettos”. These
labels have since been replaced by the less openly yet still stigmatising terms
“parallel societies and “transformation areas”, without however incurring any
substantive changes to the scheme. A neighbourhood classifies as a parallel
society when it satisfies at least two out of four socio-economic
criteria—relating to unemployment, education levels, criminal convictions, and
average gross income—and at least 50 per cent of its residents are “immigrants
and their descendants from non-Western countries”. Where such a classification
persists for five consecutive years, the area is classified as a
“transformation area” and becomes subject to the abovementioned far-reaching
restructuring measures. The Law on Public Housing also identifies “vulnerable areas”,
neighbourhoods that meet the same socio-economic criteria as parallel societies
but do not have a majority of “non-Western” residents. These areas are not
eligible for restructuring. Hence, the possibility of restructuring turns
explicitly on the “Western”/“non-Western” distinction, making the risk of
forced relocation contingent upon the demographic composition of a
neighbourhood.
The applicants are residents of designated “transformation areas” in the municipalities of Slagelse and Copenhagen. The applicants from Slagelse challenged the termination of their leases, whereas the applicants from Copenhagen sought invalidation of the ministerial decision approving the development plan for their neighbourhood. Both cases reached the High Court of Eastern Denmark, which referred two preliminary questions to the Court of Justice First, whether the criterion “non-Western immigrants and their descendants” constitutes a distinction based on ethnic origin within the meaning of the Race Equality Directive (RED); and second, whether the Danish legislation gives rise to direct or indirect discrimination.
The Judgment
The Opinion
of the Advocate General (Ćapeta) answered both questions in the
affirmative, concluding that the distinction is based on ethnic origin and that
the difference in treatment amounts to direct discrimination, although it could
also be qualified as indirect discrimination. While the Grand Chamber does not
fully adopt the AG’s Opinion, much of its reasoning is reflected in the
judgment. The CJEU, first of all, concurs with the AG that a public housing scheme
such as that existing in Denmark is capable of falling within the material scope
of the RED (§ 66). It then proceeds to examine whether the Danish legislation
constitutes direct ethnic discrimination. While leaving it to the referring
court to ultimately decide this issue, the Court offers detailed guidelines which
allow very little room for a negative answer. Still, in case the referring
court should conclude that there has been no direct discrimination, the CJEU
also provides guidelines to determine whether there has been indirect discrimination.
Here the Court interprets the requirement of Article 2(2)(b) RED that persons
of a racial or ethnic origin must be put at a ‘particular disadvantage’, as
well as the criteria for objective justification. The Court leaves to the
referring court to decide, notably, if the Danish legislation puts persons
belonging to certain ethnic groups at a particular disadvantage (§ 143) and, if
that is the case, if the restructuring measures are appropriate, necessary and
proportionate in relation to the public interest of promoting social cohesion
and integration (§§ 164-165).
On ethnic origin:
a step forward but not quite there yet
A central issue in the judgment is
whether the category of “non-Western immigrants and their descendants” falls
within the prohibited ground of racial or ethnic origin under the RED. The
Court of Justice has long displayed an uneasy relationship with the concepts of
“race” and ethnic origin — most notably with the former, which it has
traditionally been reluctant to invoke expressly. In CHEZ, the
Grand Chamber held that ethnic origin ‘has its origin in the idea of societal
groups marked in particular by common nationality, religious faith, language,
cultural and traditional origins and backgrounds’ (§ 46). While this definition
in itself is not unnecessarily restrictive, its application in subsequent cases
demonstrated a very restrained understanding of the concept. In Jyske Finans as well as Land Oberösterreich v KV the Court’s
focus was on establishing the presence of objective characteristics rather than
identifying social processes of racialisation (here) and othering through
which such characteristics gain social meaning and become markers of racialized
social hierarchy. Moreover, it was suggested that the requirement of a
‘particular’ disadvantage in Article 2(2)(b) RED meant that only ethnically
homogeneous groups could seek protection under the Directive, thus excluding
differential treatment of heterogeneously composed groups such as
‘third-country nationals’. This approach has been widely criticised in the
literature (for example here,
here,and
here).
Commentators have argued that the Court’s understanding of ethnic origin risked
hollowing out the protective scope of EU anti-discrimination law and failed to
account for the lived realities of racialised exclusion suffered by migrants in
particular.
Despite AG Ćapeta’s invitation
(Opinion, § 69) the Court does not expressly recognize ‘ethnic origin’ as a socially
constructed category. It confirms, however, that the term ‘racial or ethnic
origin’ in the RED must be understood in light of the International Convention
on the Elimination of All Forms of Racial Discrimination (ICERD) and of Article
14 European Convention of Human Rights (ECHR) and the case law of the European
Court of Human Rights (ECtHR). This confirms the Court’s growing tendency to
engage with external sources, especially its Strasbourg counterpart, even as EU
accession to the European Convention on Human Rights remains pending. The
reference to ECtHR case law is moreover significant as that court has
previously espoused a less rigid understanding of ethnicity, notably in the
Grand Chamber judgment in Biao v. Denmark where it held that ‘Danes of foreign origin’ were subject to
ethnic discrimination. The CJEU also recalls the ECtHR’s consistent
qualification of racial discrimination as a ‘particularly invidious form of
discrimination which, in view of its perilous consequences, requires […]
special vigilance and a vigorous reaction’ (§ 79). It thus makes clear that the
issue at stake is one of racial discrimination and that no relevant distinction
exists in this regard between the concepts of “race” and ethnic origin. Still,
it fails to draw any strong doctrinal consequences from this acknowledgement in
the form of a more explicitly social conception of racial discrimination.
A very welcome aspect of the case is the Court’s
clarification that ethnic origin—as previously defined in CHEZ—need not be determined by reference to a single or homogeneous
ethnic group (§§101-104; §§134-140). Returning to its first judgment on the
RED, Feryn
(§103), the Court clarifies that the notion of ethnic origin may apply to
broadly formulated criteria such as ‘allochtones’, ‘foreigners’ or, indeed,
‘non-Western immigrants’. Still, it continues to insist, as it did in Jyske Finans, that a single
characteristic such as nationality or country of birth, can never be indicative
of ethnic origin (§ 86). Instead, ethnic origin must always be determined based
on a combination of factors. As we argued earlier, this insistence on multiple characteristics
denies the reality of racial discrimination in which a single feature (such as
someone’s nationality, surname or skin colour, for that matter) may be the
reason for differential treatment if that feature functions, in the given
context, as a marker of “race” or ethnicity. The Court thus
keeps open a significant loophole in EU anti-discrimination law—one that Member
States may exploit by designing exclusionary measures around formally
non-protected criteria, while shielding them from scrutiny under the
prohibition of racial and ethnic discrimination.
In the
present case, however, the single characteristic requirement does not stand in
the way of a finding of ethnic discrimination. The Court is satisfied that the
criterion of ‘immigrants from non-Western countries and their descendants’ is
based on a ‘complex combination of criteria’, including country of birth, the
nationality and country of birth of the parents and whether those countries of
birth and/or nationality are considered, under Danish law, as ‘non-Western’
countries (§ 100).
On direct and indirect discrimination
A
distinction based on ethnic origin constitutes direct discrimination if it
involves a person being treated less favourable than another in a comparable
situation (Art. 2(2)(a) RED). In line with the AG’s Opinion, the Court identifies
two types of less favourable treatment that could result from the Danish
legislation. The first is that tenants of public family housing in
“transformation areas” face a higher chance of having their lease terminated
compared to tenants with a similar lease in “vulnerable residential areas”,
which are comparable to “transformation areas” in socioeconomic terms but where
“non-Western immigrants and their descendants” do not form more than half of
the population. It follows from the judgment that the risk of a lease being terminated early is, in itself, sufficient to
constitute less favourable treatment, there is no requirement that this risk
must have materialised. The Court moreover confirms what it had already decided
in CHEZ, namely that there is direct
discrimination when less favourable treatment results from a distinction based
on ethnic origin, even if among those affected there are people who do not
belong to the targeted ethnic group (§ 107). After all, tenants who are not
themselves “non-Western” immigrants face the same risk of losing their homes as
their “non-Western” neighbours. What is at stake here is a form of
discrimination by association, although this concept is not mentioned
explicitly by the Court.
Besides the
risk of having their leases terminated, the Court notes a second form of harm
that the applicants may have suffered, which is stigmatisation. Here the Court
shows itself much more deferential to the Danish courts: it is up to the
referring court to investigate if ‘the very name “transformation area”, which
for the residents of areas classified as such gives rise to an increased risk
of early termination of their lease, and which replaced the name ‘hard ghetto
area’ is, at national level, offensive and stigmatising’ (§ 126). It is
interesting that, at this point, the Court does show consciousness of the fact
that the meaning of certain terms is socially constructed and can therefore
differ over time and place, in this case leaving the Danish courts in a better
position to determine the stigmatising connotations of the notion of
“transformation areas”.
After
having thus applied a broad definition of direct discrimination, which acknowledges
both redistributive and recognitional harm, the Court moves on to examine the
possibility of indirect discrimination. According to the Court itself, this
analysis is necessary in case ‘the referring court concludes that the national
legislation at issue […] does not constitute direct discrimination’ (§ 130). While
this may at first sight seem a logical step to take, the Court’s consideration
of indirect discrimination becomes less obvious if it is taken into account
that the previous part of the judgment leaves very little to no room for the
referring court to find that there has not been any direct discrimination. Indeed, once direct discrimination is established, the analysis can end
there: discrimination is either direct or indirect, but not both, at least not
on the same ground.
This move is troubling. The
scheme at issue is plainly not one of indirect discrimination, and introducing
this layer of analysis risks generating confusion rather than clarity for
national courts. For one, if the referring court concludes that there is no
direct discrimination because, after all, the criterion concerning “non-Western
immigrants and their descendants” does not result in differential treatment
based on ethnic origin, this would raise the question of what would then be the
ethnic group (or groups) that could be put at a particular disadvantage, as
required by Article 2(2)(b) RED. This is a question on which the judgment
remains silent. The risk of confusion is further exacerbated by the fact that
the Court’s own conceptualization of indirect discrimination is fraught with
difficulties, as illustrated most clearly by its case law on religious symbols
and headscarves (see here
and here).
On the upside, the judgment does
provide valuable clarification as to how the case should be assessed if viewed
through the lens of indirect discrimination. This may prove practically
significant, as experience shows that even where direct discrimination is
formally removed, discriminatory effects often re-emerge—consciously or
unconsciously—through ostensibly neutral criteria. One can easily imagine, for
instance, that the Danish legislator might in the future abandon the explicit
reference to “non-Western immigrants and their descendants”, while continuing
to target specific neighbourhoods through socio-economic indicators that would,
in practice, place that very group at a particular disadvantage. Some key
takeaways from The Court’s analysis are: 1) that the aim of “ensuring
successful integration of third-country nationals (TCNs)” is legitimate in
principle; 2) the potential of the Danish scheme to actually promote social
cohesion and integration is viewed as doubtful, not least because it does not
apply to “vulnerable residential areas” that are socioeconomically comparable
to “transformation areas” but without a majority “non-Western” population (§§
162-163); 3) in addition to the requirements of appropriateness and necessity,
which are expressly mentioned in Article 2(2)(b) RED, the Court examines the
measures’ proportionality stricto sensu,
something that has not commonly done before (see here and
here);
4) regarding this proportionality stricto
sensu, it must be taken into account that the right to respect for the home
is a fundamental right protected by the EU Fundamental Rights Charter (Article
7) and that, according to ECtHR case law, the loss of one’s home constitutes “a
most extreme form of interference” (§ 170).
Structural discrimination
In sum, there is a clear
suggestion that the Danish scheme would also have to be considered as
indirectly discriminatory – at least if a particular disadvantage to one or
more ethnic groups can be established. The judgment can thus be readily
qualified as an important blow to Denmark’s “ghetto” policy. Still, from a
perspective of structural equality several problematic aspects remain – we
round off this blog by mentioning two of them. The first, already raised by one
of us elsewhere,
is that the judgment entrenches the existing distinction between EU citizens
and third-country nationals whereby only the latter are deemed in need of
integration. This becomes evident in the indirect discrimination analysis,
where it is pointed out that “integration” is a legitimate public interest only
where third-country nationals are concerned (§ 151) – the subtext being that EU
citizens are by definition already integrated.
Second, the Court seems to leave
open the possibility that, under certain circumstances, “gentrification”
measures such as those imposed by Denmark might be proportionate. This
suggestion is troubling, given that those measures reflect a discriminatory
logic that runs even deeper than the
use of stereotypes in legislative documents or the harm suffered by individual
tenants who are forced to leave their homes. At the basis of Denmark’s (or
perhaps any) integration policy lies the rationality of Goldberg’s
‘racial state’ – the notion that at its core the modern state serves to protect
public order by excluding the racialised other. It follows that any
proportionality analysis cannot meaningfully abstract from the fact that the policy
is premised on the stigmatization of racialised and impoverished communities. Yet
is seems that the Court failed to recognise this, unlike in CHEZ where it expressly acknowledged
the offensive and stigmatising nature of the practice at issue (§§84 and 108).
In short, there are several
positive points to be noted about the judgment. These are, in particular, the
Court’s acknowledgement that race discrimination does not turn on the presence
of a homogeneous ethnic group, its recognition of the stigmatisation inherent
in the so-called ‘ghetto law’ and its engagement with international instruments
and proportionality in the indirect discrimination analysis. Still, the Court
stops short of addressing the central issue at stake: the systemic and
structural character of racialised exclusion. By avoiding a direct
confrontation with structural racism, the Court ultimately does not get to the
bottom of what makes the Danish ‘ghetto law’ so deeply problematic.
.jpg){kind=link}
No comments:
Post a Comment