Void at the Centre: (The Lack of) European Union Guidance on Ethnic Data

Claude Cahn1

This paper notes, on the basis of a very brief overview of data guidance at the level of the European Union (EU), that increasing EU efforts to combat racism and racial discrimination notwithstanding, the EU has not yet ventured meaningfully into providing guidance on the production of ethnic data. This is odd, in light of (i) increasing data preoccupations at the EU level in social matters generally and (ii) a history of recognition at the EU level that data is a key instrument in securing (gender) equality. This paper further notes that the silence in the area of EU guidance on ethnic data is growing more evident, due to the growth of EU guidance on data (and other matters) in social areas generally. Indeed, the lack of guidance on ethnic data matters may be having the apparently unintended but nevertheless perverse effect of communicating to Member States the message that the EU may not be very serious in its commitments to racial equality, despite repeated action aimed at tackling racism. The failure to devote sufficient attention to ethnic data may undermine EU commitments to implementing the principle of racial equality.

The Union and Racism

Since the beginning of the 1990s, there has been serious concern at an increase in racism in Europe, as well as its most pervasive (if not always demonstrable) public expression, racial discrimination. Concerns at the continuing legacy of racism in Europe have been driven by a range of unsettling facts during the 1990s, such as (to name only two) the genocidal ethnic war just beyond the borders of the European Union in Bosnia and Herzegovina, and the increasing success of explicitly racist or xenophobic political parties in the Union itself. Efforts to combat the growth of racism (or, perhaps more accurately, to extrude it from an embedded position) have been undertaken in a number of European countries, most notably in the United Kingdom, where a high level public inquiry pronounced the London Metropolitan Police services infected by "institutionalised racism", after police investigation failed to prosecute a number of white youths suspected of the 1993 killing of Stephen Lawrence, a black teenager.

The European Union institutions have repeatedly entered the fray among those working to check the growth of racism in Europe. The Union proclaimed 1997 the European Year Against Racism. Also in 1997, the European Union Monitoring Centre on Racism and Xenophobia (EUMC) was established and in 1998 the EUMC commenced activities. In early 2000, European Union governments suspended bilateral relations with Austria after a xenophobic party – and one widely held to endorse racist views – joined the Austrian government. As a result of the Austrian episode and related concerns, amendments were included in the Treaty of Nice establishing a mechanism for the suspension of certain European Union treaty rights to Member States in the event of severe breaches of human rights or rule of law standards2.

Arguably the most fundamental change in the Union relating to combating discrimination and racism has been the adoption of a series of anti-discrimination directives, adopted pursuant to the revised Article 13 of the Treaty Establishing the European Community (TEC) after its Treaty of Amsterdam amendments3. Of particular significance, Directive 2000/43/EC "implementing the principle of equal treatment between persons irrespective of racial or ethnic origin" ("EU Race Directive") introduced legal standards throughout the Union aimed at ending differential treatment based on the arbitrary criteria of race or ethnicity4. The EU Race Directive provides details as to the scope and content of laws banning racial discrimination. The Directive includes, among other provisions, the requirement of legal remedies for victims of racial discrimination through "judicial and/or administrative procedures" for the enforcement of anti-discrimination obligations "available to all"5 and the provision that in cases in which complainants "establish, before a court or other competent authority, facts from which it may be presumed that there has been [...] discrimination, it shall be for the respondent to prove that there has been no breach of the principle of equal treatment".6 The EU Race Directive also requires that domestic law impose effective, proportionate and dissuasive sanctions for violation of anti-discrimination norms. These should include "the payment of compensation to the victim"7. Directives are binding on all EU Member States8 and transposition has also been required of candidate countries.

Although a number of early hermeneutic efforts have attempted to view the EU Race Directive as specifying merely formal equality of treatment, the Directive resists such a reduction. In the first place, its name ("implementing the principle of equal treatment between persons irrespective of racial or ethnic origin") anchors in place ambitions above and beyond mere formal procedural neutrality. In its textual provisions, the Directive provides a firm basis for moving toward securing equality of outcomes by banning both "direct" and "indirect" discrimination9. The former sails to the wind of guaranteeing only procedural neutrality by defining direct discrimination as having occurred "where one person is treated less favourably than another is, has been or would be treated in a comparable situation on grounds of racial or ethnic origin".10 However, "indirect discrimination" is taken to occur "where an apparently neutral provision, criterion or practice would put persons of a racial or ethnic origin at a particular disadvantage compared with other persons, unless that provision, criterion or practice is objectively justified by a legitimate aim and the means of achieving that aim are appropriate and necessary", thus opening the way for challenges to a range of policies, practices and laws which may have a disparate (negative) impact on certain ethnic groups. One important tool for demonstrating indirect discrimination against particular ethnic groups is statistical data on the impact of policies dis-aggregated by ethnicity.

The Union and Statistical Data

While the EU is a relatively new leader in the field of combating racial discrimination, the Union is a very established player in the area both of generating statistics (an entire EU division – Eurostat – works at the regular production of statistical data), as well as of "setting benchmarks" and elaborating "common indicators" – markers against which all member States must provide data. Traditionally, data gathered under Union guidance has been predominantly in areas of European Union competence. It should not come as a surprise that an institution which began as a treaty about coal and steel has focussed for much of its lifetime extensively on data regarding matters related to trade.

However, EU data activities have become, in recent years, ever more socially preoccupied. Thus, for example, key elements of the conclusions of the Lisbon European Council in 2000, which decided upon an ambitious programme of "modernising the European Social Model by investing in people and building an active welfare state" included the goals of "promoting a better understanding of social exclusion … on the basis of commonly agreed indicators; the High Level Working Party on Social Protection will be involved in establishing these indicators;" and "establishing, where appropriate, quantitative and qualitative indicators and benchmarks against the best in the world and tailored to the needs of different Member States and sectors as a means of comparing best practice", to name only two11.

Since 2000, the Union has acted upon decisions taken at Lisbon by developing the so-called "Laeken Social Exclusion Indicators", named after the European Council at which they were approved. These provide governments with guidelines to provide data under ten "primary indicators" and eight "secondary indicators", measuring especially poverty, but also related issues such as "self-perceived health status"12. The data generated under the Laeken Indicators should provide comparability on a number of facets of social exclusion between EU Member States.

Guidance on statistical data is not only in process in relation to EU work on poverty generally, but is also increasingly a matter for individual EU divisions focussing on particular sectoral fields. Thus, for example, the EU Directorate General of Public Health is currently in the process of developing "common health indicators". According to information provided on the European Union website, the 2004 work plan sets up seven working parties on lifestyle and other health determinants, mortality and morbidity, health systems, health and environment, mental health, accidents and injuries, and health indicators. Also according to the EU website, these working parties will in the future carry out projects in their fields with respect to: identifying and developing indicators; supporting the collection of data for the indicators at the domestic level; making data available at EU level; "reporting the knowledge deduced from the indicators" and other information and promoting the use of the results at EU level13.

EU data action on social issues has already proceeded extensively into the particular concern of equality. Equality, for the purposes of EU data efforts to date, however, is almost exclusively gender equality. Indeed, gender targets appear in very high level EU policy decisions. For example, in the field of education, the European Council has set five European benchmarks for the improvement of education and training systems in Europe up to 2010:

1. An EU average rate of no more than 10% early school leavers should be achieved;
2. The total number of graduates in mathematics, science and technology in the European Union should increase by at least 15% while at the same time
the level of gender imbalance should decrease; (emphasis added)
3. At least 85% of 22 year olds in the European Union should have completed upper secondary education;
4. The percentage of 15 year old low achievers in reading literacy should have decreased by at least 20% compared to the year 2000;
5. The European Union average level of participation in Lifelong Learning should be at least 12.5% of the adult working age population (25-64 age group)14.

Gender equality and data ambitions are even more explicitly linked in the very central EU policy area of employment. At the Lisbon European Council, the European Union set itself the goal of becoming "the most competitive and dynamic knowledge-based economy in the world, capable of sustainable economic growth with more and better jobs and greater social cohesion". The Council also considered that the overall aim of these measures should be to raise the overall EU employment rate to 70% and to increase the number of women in employment to more than 60% by 2010. The Stockholm European Council (March 2001) added two intermediate and one additional target: the employment rate should be raised to 67% overall by 2005, 57% for women by 2005 and 50% for older workers by 2010.

One might expect that the combined force of (i) long-term experience with gender equality policy, (ii) increasing attention generally to data in social fields, and (iii) vigorous new policy and law in the field of combating racism and racial discrimination, might together provide a sound basis for the expansion and enrichment of EU guidance on frameworks for data on the particular situation of ethnic groups and the impact of race-neutral policies on particular groups, particularly where there are clear indications that socially exclusionary forces are very frequently race-based. However, to date this has not been the case.

Ethnic Data and Ethnic Data Frameworks in the European Union

EU data frameworks to date have been noteworthy for the vacuum appearing where data measuring social exclusion impacts on ethnic groups should be. Thus, for example, the European Commission publication "European Social Statistics: Income, Poverty and Social Exclusion" includes no data whatsoever on the situation of individuals belonging to minority groups, including Roma, or the impact of socially exclusionary forces on specific ethnic groups15. The more recent Commission publication "The Social Situation in the European Union 2003", although also providing statistical data on social exclusion, is similarly silent on matters related to the impact of racial discrimination16. The Statistical Annex to the 2003 "Commission Staff Working Paper 'Draft Joint Inclusion Report'", although including data and "new" indicators going beyond the 18 social exclusion indicators adopted at the Laeken European Council in December 2001, likewise includes no frameworks for the provision of data by ethnicity17. And, at the higher end of the range, while the Lisbon objectives include specific objectives to reduce unemployment among women and the elderly, there is no similar objective to bring down unemployment among excluded ethnic groups.

Ethnicity occassionally slips into EU data and impact considerations via the gender gate, for example when EU policies note the need to pay particular attention to the needs of women from minority groups. Here some of the language is revealing, particularly in the apparent anxiety in some corners of the EU of even talking about ethnicity. Ponder, for example, the following formulation, from a document defining objectives in the fight against social exclusion: "the social integration of women and men at risk of facing persistent poverty, for example because they have a disability or belong to a group experiencing particular integration problems" (emphasis added).18 It is not easy to be optimistic about the prospects of success of policies to combat the exclusion of particular ethnic groups where those are designed by an institution which appears to fear giving voice to the possibility of ethnicity.

To date, a distinct tremulousness in the area of ethnic data notwithstanding, the Union has undertaken several key steps in the field of ethnic data. First of all, it has specified that generalised data disaggregated by ethnicity is not an area covered by EU data protection rules. Most recently, Regulation (EC) No 45/2001 of the European Parliament and of the Council of 18 December 2000 "on the protection of individuals with regard to the processing of personal data by the Community institutions and bodies and on the free movement of such data" states: "(a) 'personal data' shall mean any information relating to an identified or identifiable natural person hereinafter referred to as 'data subject'; an identifiable person is one who can be identified, directly or indirectly, in particular by reference to an identification number or to one or more factors specific to his or her physical, physiological, mental, economic, cultural or social identity".

Simply setting aggregate ethnic data outside the scope of data protection laws may seem like a negligible step (and it is not much really), until one becomes aware that the primary argument brought by authorities as to why ethnic data is so scarce is that gathering ethnic data may violate data protection laws. For example, in its response to the Opinion of the Council of Europe's Framework Convention on the Protection of National Minorities Advisory Committee, the German Government stated that "Germany could not consider collecting any such data due to basic legal considerations."19 As a number of articles in this issue of Roma Rights make clear, this fig leaf is worn by the governments of a number of countries, particularly but not only in Central and Eastern Europe.

A second step taken by some divisions in the EU recently has been to begin telling Member States governments that they are free to – and should – gather and make public data on the situation of particular ethnic groups. In Directorate General Employment and Social Affairs, this has been refered to as encouraging the development of "tertiary" indicators (i.e., those developed at the level of the Member State), in the absence of "primary" and "secondary" indicators (agreed upon at the EU level, and therefore refered to as "common indicators"). In so doing, the EU joins a number of international monitoring bodies which have both generally called for the provision of data on the situation of weak ethnic groups, as well as repeatedly called on states – including European Union Member States – to provide statistical data on the situation of ethnic groups in various sectoral fields.20

However, it is not at all apparent that this stop-gap measure is sufficient. In the first place, it is not clear why for example Italy should comply with a request that France won't. Secondly, with data requirements from the EU increasing now at a dramatic rate, it does not seem likely that, absent other incentives, states will voluntarily add to their data production load. Finally, the central idea of the common indicators is to provide for data comparability between Member States. Without such comparative data, it will be significantly more difficult to know which policies work and which ones do not. Or at least, it will not be possible to know which policies work better than others.

The View from the Member States

The primary obstacle to the production of ethnic data right now is that in the current conditions, the data generated will be embarrassing to those governments not yet willing to address the issues concerned seriously. This is evident already from responses to data generated as a stop-gap effort by non-governmental organisations and sociologists. To take only one example, the ERRC research conducted on the situation of Roma in the Czech school system in the school year 1998-1999 documented extreme levels of racial segregation in Czech schools21. Intensive research was carried out in the Czech city of Ostrava. The ERRC secured from school directors (who are perfectly aware of the ethnic composition of schools and who in a number of instances already have such data on file) statistical data on the placement of pupils, by ethnicity, for the approximately 90 primary schools in the Ostrava municipality. This revealed that, during the 1998-1999 school year:

  • More than half of the student body of so-called "remedial special schools" for the mentally disabled were Romani;
  • More than half of the population of Romani children of the age of mandatory school attendance in Ostrava were being educated in remedial special schools;
  • Any given Romani child was more than 27 times more likely than a non-Romani child to be schooled in a remedial special school.

These are massive racial disparities. The ERRC also conducted similar research in a number of other Czech municipalities, and this research indicated that disparities were similar or, indeed, worse, throughout the Czech Republic.

In the intervening five years since this data was first made public, the Czech government has on a number of occasions had the opportunity to present its efforts to address the problem of racial segregation in schooling, and in particular disproportionate placement of Romani children in schools for the mentally disabled. Recent Czech government pronouncements indicate an increasing awareness that the issue of racial segregation in the school system is a problem, but as yet no basis for any assessment as to the impact of policy measures adopted. Here, for example, is the Czech government's discussion of the issue in its December 2003 "Joint Memorandum on Social Inclusion", prepared with the European Commission: 

Groups with multiple disadvantages are the most vulnerable. Unemployment is high among the Roma. However, its levels can be only estimated, as no special ethnic origin-based statistics are available. [...] The unsatisfactory situation in educating children of the Roma community, who represent the majority of pupils from socio-culturally disadvantaged backgrounds, remains the permanent target of criticism from the inspection bodies of international human rights agreements, regular European Commission reports and local and international non-governmental organisations monitoring human rights compliance. They criticise especially the fact that a large proportion of them attend special schools, which de facto – although no longer de jure – limits their chances of attaining higher education levels and limits the possibility of achieving social integration. In vocational training these pupils are primarily directed towards obtaining qualifications for manual occupations. The percentage of Roma pupils attending special schools cannot be determined precisely and the existing qualified estimates cannot be generalised for the whole of the country22.

The Czech government enters into more detail on the issue of disproportionate rates of placement of Romani children in schools for the mentally disabled in its May 2003 report to the United Nations Committee on the Elimination of Racial Discrimination (CERD).23 Here, although a range of policy measures are discussed, it is again evident that no basis exists to determine whether any of these measures are having any positive impact. Absent the presentation of data on the impact of these policies, there can be no clear assessment of their efficacy or, indeed, if any of the measures described have any true life outside the confines of the Czech government's reporting to international bodies.

The Czech example is only one among many, but it is particularly illustrative, especially when data gaps are examined in light of the existing EU data framework. Precisely in the field of education, the current EU data framework seems especially prone to misread issues related to ethnicity. There are 29 "Indicators for Monitoring Performance and Progress of Education and Training Systems in Europe".24 At least two of these may actually register racial segregation of Roma in schools for the mentally disabled as a positive measure. These are indicators 4 and 5 under the heading "Investments in Education and Training" which measure expenditures per pupil/student. Since in some Member States (including in the Czech Republic), expenditures for pupils in schools and classes for the mentally disabled are higher than those for pupils in mainstream education, racial segregation into substandard, inappropriate and inadequate schooling would, in the absence of other information, register under the present indicator system as a positive measure for Roma.


The focus of this article has been the EU data framework, although the failings described above have been the failings of a particular Member State to provide adequate data on an ethnic basis. One challenge to the thesis of this paper might be, then, that the paper places pays attention to the EU level, when a focus on Member States' frameworks might be more relevant. It has not been the intention of this paper to facilitate any interest in relieving the primary obligation of Member States to provide such data. It has been the sole aim of this paper to note that a void currently exists in the broader framework, and that in light of an expansion in other areas, this void is becoming increasingly noticeable and is having increasingly pernicious effects. Said differently, there is an increasingly powerful EU framework underpinning the Czech government's current (and traditional) line of defence of policy inertia cited above: "The percentage of Roma pupils attending special schools cannot be determined precisely and the existing qualified estimates cannot be generalised for the whole of the country."25 The absence of data (and the studied unwillingness to gather it) is constant; the framework justifying it strengthens.

Among the most compelling reasons for elaborating an EU framework for ethnic data is that many governments are currently doing themselves and the societies they govern a very significant disservice. Gaps in race-specific statistical data deprive policy makers of a clear and justifiable basis for action. Data, particularly where such data indicates serious disparity, can be among the most powerful tools available for ensuring policy efficacy, facilitating social change and guaranteeing race equality. Without such statistical data, little concrete information exists which might motivate change. Perhaps most importantly, it can assist members of the public in arriving at an increased awareness of the societies in which they live, enriching the possibilities for democratic participation.

A number of concepts currently under elaboration aim at pushing the requirement to avoid negative impacts on particular ethnic groups to the level of the policy itself, through the inclusion of regular assessments and adjustments to policy. Ever more will be heard in coming years about currently misunderstood approaches such as "mainstreaming".26 The European Union can and should facilitate increasing attention to the disparate impact of certain policies on ethnic groups by elaborating its data framework to include ethnic data in all areas. This can only assist those (often embattled) policy makers seeking to combat racial discrimination, and will result generally in pressure on governments to tackle issues of exclusion on an ethnic basis. The intrinsic good secured finally is the cultivation of vibrant, multi-cultural societies characterised by social peace and regularly enriched by inter-ethnic understanding.

A new ERRC report:
Segregated Schooling of Roma in Central and Eastern Europe

The new ERRC report explores how Romani children in Bulgaria, Czech Republic, Hungary, Romania and Slovakia are denied equal dignity by a system of raciallybased
segregation in education.

Order ERRC reports at: office@errc.org


  1. Claude Cahn is Programmes Director of the European Roma Rights Center. He can be contacted at: ccahn@errc.org.
  2. Article 7 of the Treaty of the European Union, as amended by the Treaty of Nice.
  3. Beginning in 2000, and in particular under expanded powers provided by an amended Article 13 of the Treaty Establishing the European Community, the European Union adopted a number of legal measures which have significantly expanded the scope of anti-discrimination law in Europe, notably three Directives: (i) Directive 2000/43/EC “implementing the principle of equal treatment between persons irrespective of racial or ethnic origin” (“Race Directive”) (ii) Directive 2000/78/EC “establishing a general framework for equal treatment in employment and occupation” (“Employment Directive”) and (iii) Directive 2002/73/EC “on the implementation of the principle of equal treatment for men and women as regards access to employment, vocational training and promotion, and working conditions”, providing an increased level of protection against discrimination based on sex and amending an earlier directive in this area. In addition to the Directives adopted under Article 13, a revised Article 29 of the TEC now gives police and judicial authorities heightened powers to co-operate on matters related to, among other things, “preventing and combating racism and xenophobia”.
  4. The full text of the European Union Race Directive is available on the Internet at: http://europa.eu.int/comm/employment_social/fundamental_rights/legis/legln_en.htm.
  5. EU Race Directive Article 7(1).
  6. EU Race Directive Article 8.
  7. EU Race Directive Article 15.
  8. Where Member States have not transposed elements of similar directives in the past, the European Court of Justice has applied the provision at issue directly.
  9. EU Race Directive Art. 2(2)(a).
  10. EU Race Directive Article 2(2)(b).
  11. Presidency Conclusions, Lisbon European Council, 23 and 24 March 2000, pt. 37.
  12. See European Union Social Protection Committee, “Report on Indicators in the field of poverty and social exclusion”, October 2001.
  13. See http://europa.eu.int/comm/health/ph_information/indicators/indicators_en.htm.
  14. Commission of the European Communities, “Commission Staff Working Paper: Progress Towards the Common Objectives in Education and Training: Indicators and Benchmarks”, Brussels, 21.1.2004, SEC(2004) 73.
  15. European Commission, Eurostat, “European Social Statistics: Income, Poverty and Social Exclusion”, Detailed Tables, European Communities, 2000.
  16. European Commission, Directorate General for Employment and Social Affairs, “The Social Situation in the European Union 2003”, Office for Official Publications of the European Communities, 2003.
  17. COM(2003)773 final.
  18. See European Union document, “Fight against poverty and social inclusion – Definition of appropriate objectives”, Brussels, 30 November 2000).
  19. Comments by the Federal Republic of Germany on the Opinion of the Advisory Committee on the Framework Convention for the Protection of National Minorities, Opinion on Germany (19th July 2002), reference GVT/COM/INF/OP/I(2002)008, Section III, Re Art. 3, number 75.
  20. For a summary of such calls for statistical data, see Goldston, James A., “Race and Ethnic Data: A Missing Resource in the Fight against Discrimination”, in Ethnic Monitoring and Data Protection: The European Context, Budapest: CPS Books, Central European University Press – INDOK, 2001.
  21. See European Roma Rights Center, A Special Remedy: Roma and Schools for the Mentally Handicapped in the Czech Republic, Budapest, 1999.
  22. Ministry of Labour and Social Affairs of the Czech Republic and European Commission Directorate General of Employment and Social Affairs, “Joint Memorandum on Social Inclusion of the Czech Republic”, December 18, 2003.
  23. See CERD/C/419/Add.1 23 May 2003, pts. 119-128.
  24. The 29 “Indicators for Monitoring Performance and Progress of Education and Training Systems in Europe” can be found in Commission of the European Communities, “Commission Staff Working Paper: Progress Towards the Common Objectives in Education and Training: Indicators and Benchmarks”, Brussels, 21.1.2004, SEC(2004) 73.
  25. Ministry of Labour and Social Affairs of the Czech Republic and European Commission Directorate General of Employment and Social Affairs, “Joint Memorandum on Social Inclusion of the Czech Republic”, December 18, 2003.
  26. See for example “Mainstreaming Equality in European Union Law and Policy-Making”, a report prepared for the European Network Against Racism (ENAR) by Jo Shaw, Professor of European Law, University of Manchester and Senior Research Fellow at the Federal Trust, March 2004, forthcoming by ENAR, July 2004. 

donate now

Challenge discrimination, promote equality

be informed

Receive our public announcements Receive our Roma Rights Journal

news portal

The latest Roma Rights news and content online

join us

Become a part of the ERRC's activist network in Europe