About ECRI


Other CM Decisions

Internal Rules of Procedure

ECRI members

ECRI observers

ECRI in short



Country Monitoring Work

Work on General Themes



Annual reports 



Search (New HUDOC-ECRI database)

Press Releases









Restricted access

Access members

Password reset (expires every two months)


10 years of combating racism

Palais de l’Europe, Room 1
Strasbourg, Thursday 18 March 2004

Workshop II

“Racism in migration: Where do we draw the line?”

Paper prepared by Ms Isabelle CHOPIN, Deputy Director, Migration Policy Group

The views expressed are those of the author only.


During the past decades, the development of international migration policies was at European as well as at national levels removed from the socio-economic policy agenda. From the eighties onwards Europe saw itself confronted with unsolicited and undesired immigration as the great majority of immigrants entered the EU through family reunion, as refugees or irregular migrants. Europe responded accordingly by taking restrictive measures. Civil society organisations continued to call for a positive approach to migration; one based on human rights considerations and socio-economic needs assessments. Eventually labour market mismatches and demographic projections combined succeeded in re-instating economic migration on the agenda linking immigration with achieving socio-economic goals1. While immigration is re-entering the policy debates on Europe’s future it is increasingly recognised that a well-designed system for immigration needs assessment and immigrant selection is indispensable. Consultations are needed that provide for the continuous participation of stakeholders (among them those who work against racial and other forms of discrimination) at all stages of the policy process: in deciding what are the needs, how to recruit the right persons to meet those needs, and how to provide the integration efforts that must accompany any sustainable immigration policy. The issue remains equally relevant when dealing with asylum and family reunification. There are hard choices to be made because it is about including and excluding people and there are conflicting interests to be reconciled. This paper discusses the problematic relationship between selection and (non-) discrimination. It maintains that anti-discrimination norms applied to citizens and other persons within the jurisdiction of the state should also be applied to policies and practices governing immigration.

Selection and discrimination

Under international law, states are under the obligation to accept the (re-) entry of their own citizens, but they may or may not grant admission to non-citizens. Although this principle is universally accepted, it does not mean that states have an absolute sovereign control over their borders. Principles of international law and bi-lateral and multilateral treaties limit their sovereignty considerably. This in particular applies to asylum and family reunification. Furthermore, there are examples of states sharing their sovereignty also on matters of border control and immigration (the best, but not only example is the European Union). With these restrictions in mind we can say that states are indeed entitled to select the persons who may enter their territory and to remain there on a temporary or permanent basis. In other words, states are entitled to and indeed do make distinctions between persons, or discriminate in the most basic meaning of that word. In this sense migration and discrimination are inextricably linked. Although discrimination is not necessarily unlawful, the distinctions made in the design and implementation of migration policies can result in unlawful discrimination. In order to establish whether a selection of immigrants results in unlawful discrimination, one should raise the question of what the selection criteria are, how they are established and whether they are justified and proportionate. The selection criteria used in practice range from human rights’ and humanitarian commitments, interests and obligations derived from economic or historical ties between countries - i.e the racial, ethnic or national origin of persons - to characteristics of individual persons such as family ties and special skills – i.e. the social origin of persons. These criteria should be clear and transparent so as to avoid unjustified discrimination. Ideally immigration policies are a well-balanced mix of economic interests, social goals and human rights’ commitments. Refugees and asylum are matters governed by an international human rights regime that guarantees a universal right to seek protection against persecution irrespective of racial, ethnic or national origin. The situation is different as far as family reunion is concerned. Family reunion is mainly considered to be a fundamental human right more than being an element in a selective immigration regime. This may well change now that governments realise they are not in control of a major source of immigration and start designing selective immigration strategies. Proposals to control and restrict immigration will, therefore, also effect family reunion (as can be witnessed in some European countries and at EU level). The design of a new immigration regime based on selection may lead to clashes with the refugee regime (the desired immigrants versus the undesired refugees) and with the old immigration regime of mainly family-based immigration (the desired immigrant and the undesired or less desired spouse and children).

Anti-discrimination standards

Equality and anti-discrimination are enshrined in international and European conventions to which most European countries are party. These conventions inspired national policies to combat all forms of discrimination. Racial and ethnic discrimination, never totally eliminated in Europe, resurfaced at a time when European societies were becoming increasingly diverse as a result of continuous intra- and extra-European migration. Immigrant integration, of which equality formed the cornerstone, began to rank high on the national and European policy agendas. This partly explains why new anti-discrimination standards were developed2, supervisory mechanisms were established or reinforced3 and Europe-wide campaigns were held4. It is interesting to note that none of these new legal standards have addressed head on the issue of immigration and anti-discrimination. On the contrary, the EC Directive 2000/43 explicitly states in its recital 13 and article 3.2 that any direct or indirect discrimination based on racial or ethnic origin as regards the areas covered by the Directive is prohibited, prohibition which also applies to third country nationals. Nevertheless the directive does not cover differences of treatment based on nationality and is without prejudice to provisions and conditions relating to entry and residence of third country nationals and their access to employment and occupation.

To put it boldly and bluntly: whereas it is forbidden to put up a sign on public establishments with ‘no Roma, Irish or Jews’, it seems possible to close or keep countries closed for categories of people. In an increasingly borderless world with many countries of diverse populations this is an untenable situation. Therefore, we need to learn to apply anti-discrimination standards to immigration.

When speaking of discrimination, it is important to make a distinction between direct and indirect discrimination and between policy design and policy implementation. Direct discrimination is usually defined as a less favourable treatment in a comparable situation, whereas indirect discrimination occurs when an apparently neutral practice (or provision or criterion) has a discriminatory effect. For example, it is direct discrimination when a person is refused entry at the border because he or she is black or a Muslim. Indirect discrimination can be ‘hidden’ behind noble motives and lofty goals, such as, for example, immigrant integration. Persons can be required to speak the national language before receiving a residence permit; the rationale behind it could be that knowledge of the language would enhance integration into the receiving society. This could, however, amount to indirect discrimination. Immigration policies are ideally adopted after open and public debates and they lay down general guidelines and concrete measures. These measures are then carried out by a range of persons and organisations under varying degrees of public scrutiny. In both instances direct and indirect discrimination can occur, but it would seem that direct discrimination occurs more frequently in the implementation than in the design of policies. Different types of discrimination in different stages of policy formation and implementation require different skills to detect it and different strategies to fight against it.

Anti-discrimination norms regulate relations between the state and citizens (and persons within the jurisdiction of the state) and relations among citizens. This paper looks at the former and not at the latter as it tries to identify discrimination in the formulation and implementation of immigration policies. It looks in the first section at discrimination on similar and overlapping grounds of nationality, race and ethnicity. Since there is also an overlap of these grounds with culture the paper also looks at culture and in that context at language and religion. In the second section the paper will look at implementation issues and discuss what is known as racial and cultural profiling.

I. Immigration policies and discrimination on grounds of nationality, race, ethnicity and culture

1.1 National preferences in immigration policies

Immigration policies of most countries contain many elements of national preferences, resulting often from historic and economic links between countries. It is the most common and as said most accepted form of discrimination in immigration. In a way immigration policies could be seen as the management of national preferences. There are many examples of bi-lateral and multi-lateral treaties which define under what conditions nationals of one or more contracting parties can enter and settle in other contracting party.

A good example of this kind of treaty establishing a preferential treatment on the basis of nationality is the co-operation in the European Union5. The establishment of a common market without internal borders and of a political community under Community law that protects fundamental rights changed intra-Community migration into internal mobility. Or to put it differently nationality was in many cases eliminated as a criteria of the national immigration policies of the EU Member States. Community action in the field of free movement has become steadily relevant for an increasing number of people. First, through the expansion of movement rights to categories other than workers such as students, pensioners and other non-active persons who also acquired residence rights, provided that they are not dependent on the receiving state for their income and possess adequate health insurance. The second extension has taken place through the successive enlargements of the Community from six to fifteen and soon 25 Member states which will be completed in due time. It is nevertheless true that in practice these free movement rights are much more extended for EU citizens than for third-country nationals. All other third-country nationals (undocumented or with short term residence permits) are excluded from the free movement principle.

Saying that the free movement of persons in the EU depends upon nationality (or cultural preference) does not necessarily mean that it amounts to unjust discrimination. The ‘special legal order’ as the EU has been called, may justify this special treatment. However, if the EU, as far as borders are concerned, can be considered as a common market, how can the different treatment of legally resident third-country nationals be justified?

Another problematic area in EU law, with regard to discrimination in migration, concerns the Schengen system. This system puts states under the obligation to deny access to individuals who have been put on the Schengen ‘black list’ by any of the other participating states. This means that such an individual may be on that list for reasons that would not have lead to exclusion in another Schengen country (e.g. membership of an organisation which is illegal in one Member State but not in another). Or to put it more bluntly, how can it be justified that under the Schengen system a country must refuse entry of a person who has been put on the black list by another country, whereas it may refuse the entry of a person who has been admitted to another country? From a legal point of view this is problematic because states are not obliged to make their own assessment of the refusal of entry, which is possibly discriminatory. In a recent case put before the European Court of Justice, the Commission has asked to rule on such a negative decision by Spain.6 A family member of an EU citizen was denied access to Spain for the sole reason that Germany had put that person on the Schengen black list. The question is whether this list is sufficient evidence that the person involved constitutes a threat to the national security of Spain. The European Commission finds that the ‘public policy exception’ of regular EC law does not fit with the exception in the Schengen agreement and Member States would have to make their own assessment.7 It is now up to the Court to decide whether this is an illegal and discriminatory practice or whether it is justified by the ‘special nature’ of the Schengen co-operation.

Although it is true that for long-term residents new legislation has entered into force that will give them a limited free movement right in the EU, such legislation does not yet exist for persons with short term residence rights. Here the term ‘cultural profiling’ no longer applies, but it certainly makes a case for indirect racial discrimination. Indeed, the rules may seem neutral on the face of it, but they concern in practice especially immigrants from Turkey and the Maghreb countries.

Other countries apply similar preferential treatment with regard to certain countries. Thus in Estonia, the quota for immigration does not apply to citizens of the EU Member States, Norway, Iceland, Switzerland, Japan and the USA. Similar legislation also exists in Switzerland and Liechtenstein.

1.2. Race-based immigration policies

In the so-called traditional immigration countries Australia, Canada and the United States policies were, for a long time, based on national origin of immigrants. These systems had clear racialist overtones and were therefore abolished. In the US this took place in the 1960’s and was an indirect effect of the civil rights movement. Around the same time Australia gave up its ‘white Australia’ policy and Canada introduced its ‘multicultural policies’ and applied it to immigration. When opportunities and entitlements are granted irrespective of race, ethnicity and national origin, it is difficult to defend a position that this only applies to those who are already present and not to those who want to come in. Whereas immigration policies will probably not explicitly exclude persons on the basis of their racial origin, they could prefer persons belonging to an ethnic or racial group. Examples are policies in Germany (Aussiedler), Finland (Ingrias), Greece (Pontians) and Israel (Jews). These types of policies could be qualified as ‘affirmative immigration policies’ or as ‘positive discrimination in immigration’. In Europe, affirmative action and positive discrimination are generally being considered as not acceptable and going against equality and anti-discrimination principles. At best these preferences in immigration can be called ‘positive action in immigration’. The justifications for these preferences are to be provided by law. The preferences must serve a defined purpose and be subject to review and be temporary.

2. Cultural preferences in immigration policies

There is a fine line between discrimination on the basis of race and ethnicity and discrimination on the basis of culture. Europe seems to be moving towards maintaining a strong position versus racial and ethnic discrimination and a softer position versus cultural discrimination. Immigration policies based on race would cause an outcry, immigration policies based on cultural considerations is becoming debatable. Such policies would include cultural criteria in the selection of immigrants and would make such criteria decisive. They would give preference to immigrants originating from countries with which the host state has certain historic and cultural links. Obvious examples are countries with (historical) colonial links with other countries, or countries which once formed one country or were part of the same empire or groups of states. It is claimed or suggested that there would not be obstacles for the integration of immigrants from countries with a similar or the same culture. There are two elements of culture that are currently receiving more attention in the integration debates and which soon may play a role in the immigration debate, namely language and religion.

In a number of European countries language has become a focal point of integration policies. Language courses are becoming compulsory in these countries and immigrants who fail to learn the language will be sanctioned (for example, by withdrawing or delaying residence permits or welfare benefits). One step further and a certain level of knowledge of the national language will become a condition for admission and residence, which could even be applied to persons who are entitled to immigrate to that country (for example in the framework of family reunion). Can a spouse not being able to speak the national language of the host country be deprived of the right of family reunion? This would amount to discrimination on the basis of language. More generally, the preference for immigrants with the same linguistic background or the capacity to speak the national language is quite common, in particular when it concerns the recruitment of (temporary) workers. In this instance language requirements form part of a policy for the recruitment of workers. In such policies a variety skills (and language could be one) form an integral part of selection criteria which are objectively defined and related to labour market needs.

Making religion an immigration selection criteria is an idea that comes up regularly. Not so long ago this idea was promoted by an Italian Roman Catholic Church official who made a plea for recruiting immigrants from catholic countries. The idea had some resonance in political circles (and not only in Italy) as a possible way of barring much needed immigrants from being recruited in Muslim countries. The reasoning behind it seems to be that Catholic or Christian immigrants integrate more easily into European societies than Muslim immigrants. This is ‘substantiated’ by the perception that immigrant integration has failed and that much of this failure is attributed to the incompatibility of Muslim values and norms with Western values and norms. It is interesting to see that in some instances anti-discrimination or equality arguments are used to underpin this line of thinking (namely, discrimination against women and unequal treatment of women by Muslims). An immigration policy based on religious preferences would have to spell out which religion is to be preferred over other religions and how such preferences relate to the stated goals of integration of which equality is the cornerstone. When on the basis of recent legislation8 religious discrimination is forbidden in employment, it is difficult to see that measures would stand up in Court which explicitly exclude Muslims from being recruited as migrant workers. Equally it will not hold to deny families the right of family unity on the basis of the fact that they or some members of the family are Muslim.

II. Implementation of immigration policies and profiling

In some Anglo-Saxon countries the term ‘racial profiling’ is a hotly debated issue. Racial profiling has been described by the US Department of Justice as “any police-initiated action that relies on the race, ethnicity, or national origin rather than the behaviour of an individual or information that leads the police to a particular individual who has been identified as being, or having been, engaged in criminal activity”9. In the USA, for instance, a judgment of the Court of Appeals condemned (in 2000) the practice of border patrol agents to stop individuals for questioning based solely on their “Hispanic appearance”. This judgment was based on section 14141 of the Crime Control Act of 1994 which forbids, amongst other things, the misconduct of public officials, including discriminatory harassment and unlawful stops, searches or arrests. Although the term ‘racial profiling’ was not mentioned with so many words in that act, that is exactly what the judgment condemned. The term ‘racial profiling’ in the context of immigration has not received the same attention by policy makers and scholars in Europe as it has in some Anglo-Saxon countries. Can we assume therefore that it does not exist? And if it does exist, should we care?

The most obvious manifestation of racial profiling in the area of migration is the selection made by immigration and border officials deciding who to perform checks on. It will not come as a surprise that persons of a physically recognisable differing race, ethnic origin or religion are more likely to be asked to identify themselves than others, even if there is no objective justification for these frequent identity checks. The case of UK customs officials checking passports at Czech Republic airports and discriminating overtly against Roma passengers is a well-reported case.10 As a matter of fact, the exception for positive discrimination in the UK Race Relations (Amendment) Act 2000 was explained by the Home Secretary as to allow for discrimination against certain ethnic groups, namely Tamils, Kurds, Pontians Greeks, Roma, Somalis, Albanians, Afghans and others11.

People of colour or recognisable religious minorities are more likely to be asked either by customs officials or police for their ID or residence permit. These practices occur in all countries and on all administrative levels. Just to mention a few examples: observers have reported discriminatory checks on the part of immigration officers of non-whites coming to Cyprus; in Finland, waiting periods concerning asylum applications for Turks are longer in comparison with other nationalities, Germany performs so-called ‘pre-field control’ at certain airports whereby the passports of passengers on planes from certain destinations are checked at the exit to the plane, before the ordinary control inside the building, etc. All of these practices carry in them traces of racial discrimination.

Also with regard to the issuing of visas racial profiling may exist. Governments use lists of countries whose nationals need visas to enter their country, either for permanent residence or for short term stays or even transit. The drawing up of the list of countries is the implementation of the policy according to which citizens from some countries are subject to the visa obligation.

The way this ‘black list’ of countries with a visa requirement is drawn up almost excludes by definition countries from the Western and developed world. They include, also by definition almost always developing countries or, especially since recent events, countries that might be considered likely suppliers of terrorists or in any other way possibly harmful for national security. Indeed, whereas the criteria for requiring visas used to be based on inter-state relations, the EU employs much wider criteria. The decision whether or not to put a country on the black list mainly depends upon three considerations, varying from (1) illegal immigration to (2) public policy and security and (3) the EU’s external relations with third countries.12 In other words, a person may be subject to the visa obligation because he or she originates from a state whose nationals are, according to the statistics employed by the EU institutions, likely to enter illegally or commit crimes. The focus has shifted away in the EU from the traditional state-to-state relationships to individual-to-state relationships. This way, persons from these countries are, directly rather than indirectly, discriminated against. Cholewinski draws the conclusion that “this assessment of risk is not undertaken in respect of the activities of a particular individual but on the basis of the broad criterion of nationality.”13 It is actually a very clear case of profiling; the visa requirement is imposed on nationalities who are likely to be a risk (of illegal immigration, crime or terrorism).

Besides the questions of racial and cultural profiling, a few words also need to be said on the possibly discriminatory effects of the right to family reunification for immigrants. It is a known fact that family reunification is numerically the main reason for legal immigration in Europe. This form of immigration clearly gives preference to persons who already have family relationships in a receiving country. Whereas this fact may be discriminatory as such, it also hides a less overt form of discrimination, namely sex discrimination (far more women than men enter for reasons of ‘family formation’), discrimination on grounds of sexual orientation (if marriage is a pre-condition for family reunification, homosexuals hardly stand any chance) and discrimination on grounds of marital status (as before). But also within family reunification discrimination takes place. Here we enter again into the discourse of racial and, especially, cultural profiling. Indeed, recent measures in many European countries try to make the barrier for family reunification more difficult to pass. Countries require a high financial threshold, a more restricted age limit (children need to be younger, spouses need to be older), specific language requirements or a specific size of accommodation (established number of m2 per person) before receiving a visa for the host state.

With regard to international case-law on this issue, the Ahmut-case of the European Court of Human Rights in Strasbourg may be mentioned. In this case the Netherlands had taken a negative decision on the application for family reunification by Mr Ahmut, who had both Dutch and Moroccan nationality, with his underage son. The decision was apparently based on the fact that Mr. Ahmut was of a different ethnic origin, notwithstanding his Dutch passport. The European human rights court (deciding with no more than five votes to four) mentioned that “the fact of the applicants’ living apart is the result of Salah Ahmut’s conscious decision to settle in the Netherlands rather than remain in Morocco” and “Article 8 [ ECHR, IC] does not guarantee a right to choose the most suitable place to develop family life.” One of the Strasbourg judges, Judge Valticos, expressed a dissenting opinion. He observed that “The father had acquired Netherlands nationality, and in any country, a national is entitled to have his son join him, even if the son does not have the same nationality”. He questions the decision by the Dutch authorities not to grant the right to family reunification in this case and comes to a troubling conclusion. His statement reads: “How does it come about that in the present case this right was refused to him? I cannot think that it is because the Dutch father was called “Ahmut”. However, the suspicion of discrimination must inevitably lurk in people’s minds.”

As the following example demonstrates, the lines between racial or ethnic profiling and cultural and religious profiling are fine. In the Abdulaziz-case,14 the preference of the UK for Common Wealth citizens was the object of discussion. The European Court of Human Rights was confronted in this case with the opinion of part of the referring Human Rights Commission who had found that the legislative history of the UK immigration Act was to “lower the number of coloured immigrants.” In its judgment, the Court refers to the Commission conclusions, which recalled that most migration policies differentiate on the basis of people’s nationality, and indirectly their race, ethnic origin and possibly their colour. Whilst it is true that a Contracting State cannot implement policies of a purely racist nature, the Human Rights Commission considered that “to give preferential treatment to its nationals or to persons from countries with which it has the closest links [does] not constitute “racial discrimination”.

Thus, according to this judgment, the different forms of preferential treatment mentioned before, including racial and cultural profiling, do not amount to unlawful discrimination. That was so in 1985. One may wonder whether the Court would still come to the same conclusion today.


These examples bring us back to the beginning of the paper where the question was submitted: where do we draw the line of selection as part of national sovereignty, and where does discrimination and racism begin? In the Abdulaziz-case the majority of the Commission found that even rules which history has shown to have been drawn up to prevent the entry of “coloured immigrants” were acceptable. Would that assessment still be the same today? Is it acceptable to select on the basis of family links? On professional skills? On capital? On age? On language? On sexual orientation? On religion?

The question indeed is not so much whether states are allowed to make a selection in the immigration process, because it is generally accepted that they are, but rather on what grounds, and where do preferences become inadmissible. One could be tempted to say that preferences, profiling or discrimination in migration with regard to race or ethnicity are never acceptable, notwithstanding the historical, cultural or geographical links that may exist between the host country and the country of origin. The position according to which the links existing between the host country and the country of origin could justify a discrimination based on race or ethnicity cannot be defended from an (international) anti-discrimination point of view. With regard to the other considerations mentioned, preference on these bases is not by definition excluded, but a constant assessment should be made to evaluate if the conditions are not a cover-up for forms of indirect discrimination. Thus, nobody would advocate a system whereby no account is taken of family relationships.

As was the case when drawing up anti-discrimination legislation, public debate and dissemination of information seems crucial. Indeed, more restrictive policies and practices are often vouched by public opinion prejudiced against migrants (whether they are legal, illegal, refugees or asylum seekers).

It is the task in the first place of policy makers and democratic institutions to keep watch over the quality and the motivations of migration policies. But civil society, in the form of NGOs or other interest groups, should be called upon to watch and correct the guards where necessary.

1 See Jan Niessen and Yongmi Schibel, Demographic changes and the consequences for Europe’s future. Is immigration an option (MPG, 2002 – www.migpolgroup.com).

2 The three main new instruments are Protocol XII (which has not yet entered into force) to the European Human Rights Convention and the two EC Equality Directives 2000/43 and 2000/78. See Jan Niessen and Isabelle Chopin, “the developments of legal instruments to combat racism in a diverse Europe”, Immigration and Asylum Law and Policy in Europe, volume 6, Martinus Nijhoff publisher 2004.

3 Such as the European Commission against Racism and Intolerance (Council of Europe) and the European Monitoring Centre on Racism and Xenophobia (European Union).

4 Such as the Council of Europe campaign “All different, all equal” and the EU 1997 European Year against Racism.

5 And on a much more modest scales in terms of seize and scope in the Nordic Labour Market or the Common Travel Area between the UK and Ireland.

6 Case C-503/03.

7 For more details, see Migration News Sheet February 2004.

8 The EC Directive on equal treatment in employment. See for the transposition of this Directive in current EU Member States the website of the EU Commission (DG Employment and Social Affairs) and in accession states: MEDE and the Migration Policy Group, Equality diversity and enlargement. Report on measures to combat discrimination in acceding and candidate countries (DG Employment and Social Affairs, 2003).

9 Deborah Ramirez, Jack McDevitt, and Amy Farrell, A Resource Guide on Racial Profiling Data Collection Systems: Promising Practices and Lessons Learned (Washington DC, Dept of Justice, November 2002).

10 News paper coverage between July and August 2001 (e.g. The Independent and The Guardian).

11 Ethnic Chinese presenting a Malaysian or Japanese passport or any other travel document issued by Malaysia or Japan. For more details, see R. Cholewinski, Borders and discrimination in the European Union, ILPA/MPG 2002.

12 Fifth recital to the Preamble of the Visa Regulation.

13 R. Cholewinski, Borders and discrimination in the European Union, ILPA/MPG 2002.

14 Abdulaziz, Cabales and Balkandali v. UK (1985) 7 European Human Rights Reports [EHRR].