There is a large variety of policies and actions contributing to tackling discrimination against immigrants and ethno-racial minorities. These policies can be distributed along a gradient from formal equality to proactive policies that could include preferential treatment for disadvantaged groups. Antidiscrimination laws and policies aim to prevent negative and unjustified distinction, exclusion, restriction, or preference based on grounds such as nationality, race, color, sex, language, religion, political opinion, etc. The list of grounds varies across countries: the French law, for example, identify no less than 25 criteria of discrimination, the law in countries such as Denmark or the UK operates with eight criteria, while the German General Equal Treatment Act (2006, amended 2013) mentions only six grounds. A large number of countries have chosen an open-ended list to avoid restricting the scope of discrimination.

Antidiscrimination laws and policies aim to ensure equal rights for the protected groups (e.g., women, people with disabilities, or ethnic and racial minorities). The main goal of these legal provisions, policies, and actions is to achieve equality for all in concrete terms and not only in principle. According to De Witte (2010), the common principle of equality is “broad and empty” and should be specified to become substantive. Fredman assigns four objectives to such substantive equality policies: “to redress disadvantage; to address stigma, stereotyping, prejudice and violence; to enhance voice and participation; and to accommodate difference and achieve structural change” (Fredman 2016, 713). However, while the principles and objectives of equal rights, equal treatment, and equal access to resources, goods, and services receive generally large support among policy makers and public opinion, concrete positive actions tend to be more divisive. This is especially the case of positive discrimination, which provides preferential treatment – an advantage – to members of protected groups to redress the penalties they historically have faced (and often still face), in access to higher education, political mandate, public jobs, or social housing.

Importantly, countries vary greatly in their strategies to tackle ethnic and racial discrimination. First, they can be divided into two groups: those who have adopted ethnic and race-based policies, or ethnic and/or racial conscious policies, and those who favor color-blind policies, meaning that they address ethno-racial discrimination without identifying explicitly categories of victims based on ethnicity and race (see Chap. 1). Second, they diverge in the kind of measures they implement in the name of antidiscrimination policies. There are three main groups of measures – antidiscrimination legislation, affirmative action and other equal opportunity policies, and tools for promoting diversity. The chapter discusses these different measures in turn, before turning to studies that have aimed at assessing the effectiveness of measures to combat discrimination.

7.1 Antidiscrimination Legislation

Following the Universal Declaration of Human Rights, a series of international treaties and conventions promoted by the United Nations have set international norms for equality: The International Human Rights Charter, the International Covenant on Economic and Social Rights and the International Covenant on Civil and Political Rights. Principles of equality have further been detailed in thematic conventions, some of which specifically focus on racial discrimination. The International Convention on the Elimination of All Forms of Racial Discrimination adopted in 1965 and the Convention 111 of the International Labour Organization on discrimination (employment and occupation) adopted in 1958 are the main references in this area.

In Europe, the Racial Equality Directive (RED) enacted in 2000 constitutes the main legal framework on ethnic and racial discrimination. It implements the principle of equal treatment between persons irrespective of racial or ethnic origin and complements the European directives on discrimination in employment (which covers several grounds) and other directives dealing specifically with gender, age, disability, religion, or sexual orientation. The RED came relatively late after the pioneering antidiscrimination law implemented by the UK in 1976, which served as a reference for the European Commission. Similar legislation can be found in immigration countries at much earlier dates – such as Australia’s Racial Discrimination Act of 1975, the Canadian Human Rights Act of 1977, and Title VII of the Civil Rights Act in the US enacted in 1964 (Simon 2005).

Each antidiscrimination law provides for the creation of agencies responsible for monitoring its application and for implementing its programs. At the inception of the process, agencies tend to be specialized on a specific ground (gender, race and ethnicity, disability), but the recent trend is to merge these together into a single body. For example, the British Commission for Racial Equality, the Equal Opportunity Commission, and the Disability Rights Commission were grouped together in the Equality and Human Rights Commission, established by the Equality Act of 2006. The creation of an independent equality body is a requirement spelled out in the RED, and all EU member countries have more or less complied with this. In addition to the national equality bodies, the European Commission established the EU Agency for Fundamental Rights (FRA) in 2007, as well as a network of equality bodies, called EQUINET, created in 2002–2004. However, even in the common framework provided by the EU directives, antidiscrimination actions vary greatly among EU countries. The prerogatives of these agencies in combatting discrimination can be far-reaching, ranging from the awareness raising of public authorities and civil society to the coordination of equality policies. They are responsible for all complaint-handling activities and may conduct legal actions and investigations.

Antidiscrimination laws can be enforced in civil, administrative, or criminal courts. There are important differences in these legal tracks in terms of plaintiffs, procedures, and sanctions or sentencing. However, enforcement of the law can take non-judicial procedures aside from these judicial proceedings: negotiation or mediation can be actively promoted by equality bodies that are not judicial entities. In addition, labor inspectorates are often charged to enforce the employment law and its provision on discrimination.

The legal context itself produces large disparities in the outcome of the legal actions, and differences in organizational structures have an impact on the efficiency of the legal antidiscrimination framework. Comparative studies on the implementation of antidiscrimination laws have shown significant variations across European countries when it comes to access to rights and the efficiency of legal action. For example, shifting the burden of proof – meaning that the defendant (e.g., the employer) has to prove that the treatment was not discriminatory – is not available in all EU countries, and in those where the provision exists, not in all judicial procedures. Protections against victimization of plaintiffs in retaliation of their claim are inconsistent in some countries, and lack credibility in others. Sanctions and remedies differ greatly in their capacity to punish and prevent discrimination acts, reflecting the different concepts of equality and the legal order governing each national context. Even under the EU antidiscrimination law, no comprehensive system has been adopted so far.

Equality bodies are generally entitled to receive complaints, to assist victims in litigations and sometimes have the legal power to take sanctions and make legal decisions. Negotiation, mediation, or conciliation are often preferred to litigation since discrimination cases often proved to be difficult to prosecute in the courts. Equality bodies have frequently prioritized strategic litigation whereas a limited number of cases are selected to set changes in court practices. Filing a complaint in court might be complicated in some countries, and the outcome of these complaints are rarely successful (FRA 2012). A gap between complaints and lawsuits can be observed in France where the former equality body (HALDE) treated 5658 files of complaints in 2010, of which 127 legal cases were completed (in various categories). In less than a handful of cases, condemnations actually took place, although a large number of files had been treated through mediation. In general, legal action against ethnic and racial discrimination is less developed than against sex or disability discrimination. For example, in England and Wales in 2019, the Employment Tribunal has treated 9427 complaints of sex discrimination, 6919 for disability, 3589 for race, and 753 for religion. In addition, 27,730 cases came under the equal pay law, which is a sub-type of sex discrimination. Although legal prosecution is an important part of antidiscrimination action, the legal framework has to be complemented by policies and more proactive strategies to control practices and processes without waiting for a complaint to be filed in.

7.2 Antidiscrimination Policies: Positive Action

Despite the difference in wording, affirmative action and positive action are essentially the same kinds of policies. The former concept originated in the US, while the latter, inspired by the UK, was adopted by the European action plan against discrimination (McCrudden 1986). As Daniel Sabbagh summarizes it, the goal of such positive action “is to counter deeply entrenched social practices that reproduce group-structured inequality (even in the absence of intentional discrimination) by creating positive externalities beyond individual recipients” (Sabbagh 2011, 109). Still, there exists a variety of measures in positive action policies that differentiate them along a continuum of the transformative powers of the actions.

7.2.1 Awareness Raising

All antidiscrimination policies begin with awareness raising through communication campaigns. The objective is to disseminate the framing in terms of discrimination to create consciousness among victims and potential authors. Indeed, the capacity to tackle discrimination depends on the conceptualization of the phenomenon, as well as the underlying understanding of how it operates and what consequences it causes for disadvantaged groups. There are different ways to address biases and inequalities generated by discrimination, beginning with programs to empower underrepresented minorities, actions to pursue a higher level of impartiality in decision-making by acting directly on processes and developing training and eventually schemes to impose preferential treatment for certain categories of disadvantaged groups, including quota systems. In the following, we detail some of these actions with examples from practices in different countries. Although there are trends of cross-national harmonization of legal frameworks, antidiscrimination policies tend to remain country-specific. What applies to one country might not be available in another one, even in Europe where the European Commission has stimulated the adoption of common legal and practical tools.

7.2.2 Outreach Programs

One way to increase participation in the education or labor markets is to develop information about opportunities to underrepresented ethnic and racial minorities. These programs are called “outreach” because they target specific population groups or places that are usually not reached by information about the existence of opportunities. The rationale behind these programs is that minorities do not consider applying to selective tracks in education or advantageous job positions because they do not feel entitled to it or do not have access to the relevant information. Outreach programs are frequent in education to attract minority students in selective programs where they tend to be highly underrepresented. In France, for example, dedicated preparatory programs were developed in the 2000s to ease the access to elite schools (grandes écoles) for students from high schools located in disadvantaged neighborhoods (Allouch and Buisson-Fenet 2009). In employment, these schemes build on the so-called spatial mismatch theory (see Chap. 3), which suggests that minority members experience greater distance from job markets both spatially and culturally, thus attempting to compensate for this structural disadvantage by disseminating the information about job opportunities in specific locations or toward minority groups. Outreach programs aim at increasing the critical mass of minority applicants but do not address potential discrimination in selection processes.

7.2.3 Proactive Policies

One of the main goals of positive actions is to address non-intentional, systemic, and indirect discrimination by identifying biases in apparently neutral procedures. These biases are harder to identify than unfair treatment justified by the expression of prejudices. Subtle discrimination is mainly detected as their disproportionate negative consequences on protected groups. The EEOC in the US defines an adverse impact in employment as “a substantially different rate of selection in hiring, promotion, or other employment decision which works to the disadvantage of members of a race, sex, or ethnic group.” The EU law develops a similar approach in its definition of indirect discrimination (see Chap. 2), as the European Convention on Human Rights which retains 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.”

Thus, decisions, procedures, and selection schemes (in employment, housing, education, but also in the allocation of goods and services) have to be monitored to check the impartiality or neutrality of the process. Monitoring systems are frequently, but not exclusively, using statistics to detect under-representation of protected groups and biases in processes of selection or allocation of goods and services. It should be clear that the notion of fair representation is attached to those of statistical under-representation, which gives a paramount role of statistics in the identification of discrimination, the design of policies, their implementation, and their evaluation.

In order to be effective, equality programs in employment must follow these steps as part of their implementation: First, the definition and identification of members of protected groups. This is necessary to collect data, and especially statistics, on their proportion in all aspects of the employment process, such as in the applicant pools. Second, to collect data on the distribution of protected groups in different occupations in the firm, according to the level of qualification of the employees, wages, terminations, access to on the job-training, etc. Third, to compare these data to a statistical benchmark computed at different geographical levels and inside the firm itself to identify the potential gaps, which should then be corrected. Based on these statistical assessments, action plans are designed to reduce or suppress biases at the different steps of the employment relationship (hiring process, wage setting, and career advancement). In essence, equality programs combine the goals of improving the representation of protected groups with meritocratic criteria, since qualifications and skills are still the determining factors in the protected groups’ representation.

7.2.4 Quantitative Targets and Quotas

Redressing the under-representation of protected groups can be achieved through quantitative objectives. The idea is to measure the evolution of the participation of protected groups to the organizations until they reach a threshold that has been established beforehand. These quantitative objectives can be mandatory, and in this case, one can speak of quotas to achieve, or an invitation to reach a target without specific sanctions if the organization fails to meet its objectives. When a quota is imposed, the organization (university, employer, landlord, parliament) must select a number or proportion of applicants with a specific characteristic (e.g., gender, ethnicity or race, disability, religion) to be incorporated in the program. An example can be given in political representation with reserved seats for women in India or legislated gender quota among candidates to political mandate in six EU countries, in employment for people with disability or in education for ethno-racial minorities in the US in the first phase of affirmative action (until 1973 in employment and 1978 in education). If the quota is not achieved, sanctions (generally financial penalties) against employers or universities might be enforced.

The legitimacy and efficiency of quotas have been extensively discussed in the US, especially during the 1980s with the disengagement from affirmative action by the administration under President Reagan. Although the available research suggests that quotas can be an effective tool, this instrument has often been poorly implemented and remained a contentious provision that is often criticized (Stryker 2001). As a policy tool, racial quotas have been discontinued in the US, but remain in some countries such as Brazil and Malaysia.

In opposition to quotas, most of the countries have adopted a more lenient approach by setting targets and goals that are still using quantitative tools but not in a mandatory way. For example, positive actions in the UK or equal employment opportunity policy in Canada are explicitly forbidding any quota. In these cases, the advantage given to members of protected groups does not appear as explicit as it is the case for preferential treatment.

One important condition for implementing these quantitative strategies is to be able to produce statistics broken down by ethnicity or race, or any kind of relevant category under protection. When it comes to ethnicity and race, the availability of such statistics is rather limited in most of the European countries, and thus limit the diffusion of these tools.

7.3 Promoting Diversity

Aside from public policies, there are initiatives undertaken directly by the business community. Although diversity management at its inception was a by-product of equal employment policies (Dobbin 2011), it has often been implemented by companies in countries where such policies have never been developed, especially in Europe (Wrench 2007). Indeed, the spread of diversity management seems to reflect the extension of multinational companies and the standardization of human resources processes. Diversity management tools include audits to identify biases in the organizational processes, mentoring programs, career guidance, diversity training, outreach activities toward underrepresented groups to diversify recruitment channels, etc.

The main idea behind these initiatives is that creating a diversity-friendly workplace by facilitating the recruitment, inclusion, promotion, and retention of “diverse employees” and managing properly this diverse workforce will help to increase productivity and give a market advantage to companies both in the domestic market – by reaching out to immigrants and their descendants as customers – and in markets abroad. Likewise, in the context of labor shortages, developing diversity management tools has become an important means for attracting and retaining staff. In addition, there may also be a value-added stemming from diversity itself because bringing together people with different backgrounds, experiences, and perspectives may increase the potential and the expertise of the working unit. Developing a diversity plan and targeting a fair representation of minority members in the workforce also have other benefits by helping to reduce the risks of litigations. The objective here is the reduction of the legal threat and the penalties resulting from legal cases. Further, employees may favor working environments that promote inclusion, respect, openness, collaboration, and equity. Finally, diversity management may involve benefits in terms of better publicity, and thus be used as a reputational tool by the firm. The European Commission has popularized the advantages of diversity in the economy under the heading of the business case for diversity (2005).

Diversity management has its roots in the US during the 1980s, during the peak of equal employment policies. A new class of “diversity managers” was created to implement actions against systemic discrimination rather than intentional discrimination. In 1980, diversity management was applied by less than 5% of a sample of 389 employers surveyed by Dobbin and Kelly, and almost 50% of them had implemented it by 1997 (Dobbin et al. 2007). In Europe, a survey conducted in 2005 found that 52% of companies did not develop any diversity initiatives, and only 21% had well-embedded policies and practices (European Commission 2005). The main motivations of these latter companies were (1) “commitment to equality and diversity as company values,” (2) “access to new labor pools and high-quality employees,” and (3) “economic effectiveness, competitiveness, and profitability. In contrast to the US, compliance with the law was not a major driver for these companies, which reflects that the antidiscrimination framework in Europe tends to be less pressuring. Interestingly, the survey showed also that only 31% of the companies implementing diversity initiatives were monitoring and reporting the results and impacts of their actions. In the remaining 69%, enhancing diversity was mainly an intention that could not be assessed.

Whereas equal employment policies comprise legally binding compliance to standards and codes of practices, fulfilling a diversity charter or acquiring a diversity label depends on voluntary initiatives from organizations. In contrast to the latter, however, these tools involve public or semi-public bodies that are at least proposing the tool and – in the case of labels – involve certifying participation and compliance.

A diversity charter is a document by which a company or a public institution commits itself to respect and promote diversity and equal opportunities at the workplace. More or less detailed provisions or targets can be stated in these charters. One of the first of its kind in Europe, the French diversity charter, was launched in October 2004 and has been signed by more than 3450 companies since then. This example has been replicated by almost all EU countries. The country-specific charters differ by their coverage and their scope, but the commitments are similar in their principles. Being voluntary, these charters do not entail specific monitoring to check if companies respect their commitments. As such, the charters testify that the companies show some concerns about promoting diversity, even if such a concern may not necessarily translate into concrete actions. Reviews of the actions implemented according to the charter are suggested, but in most cases, the audits focus on the design of the programs and not on their outcomes.

Diversity labels go one step further by delivering a certification based on an assessment of the measures taken and their implementation. An independent body is responsible for delivering the label, which is based on an audit of the companies. A diversity label was established in France in 2008 and is delivered by a commission made up of representatives of the national administration, the social partners, the National Organization of Human Resources Managers and experts. The label is delivered for 3 years; more than 260 companies have received it thus far. A similar diversity label is granted by the Brussels-Capital Region in Belgium. Some countries, such as Belgium, have also established specific diversity awards, rewarding good practices in this domain by employers.

Among the elements that can produce discrimination, notably with respect to the crucial first stage of the recruitment process, the formatting and contents of the CV of job applicants have been a major concern among equal opportunity policy makers and diversity managers. The recruitment process involves some kind of discretion from recruiters, and the more the room for discretion, the more stereotypes and prejudice might be activated. A concrete strategy to reduce the level of discretion in hiring procedures is to standardized job application documents in a way that only useful information about the applicants should be delivered. Building on the findings of correspondence test studies that clearly show that names and other signals of minority background foster negative selection (see Chaps. 4 and 5), the idea to promote blind or anonymous CVs has gained traction in France, Germany, and the UK. The advantage of anonymous CVs is to reduce the information that conveys signals related to discrimination, such as age, gender and ethnicity/race or nationality. The expectation is that applicants who will not be screened out at the first stage of the process will be able to demonstrate their capacities at the later stage and will eventually access higher opportunities for recruitment. A body of studies has tried to measure the outcomes of this measure in Germany (Krause et al. 2012), the Netherlands (Blommaert et al. 2014), France (Behaghel et al. 2015) and in Sweden (Aslund and Skans 2012). All of these studies but one (in France) found that ethnic minorities benefit from anonymity, but still encounter a harder selection at the stage of the job interview. The French study concluded that while women did benefit from anonymity, this was not the case for applicants with a minority background. One explanation for this unexpected finding, shared by Krause et al. (2012) in Germany, is that employers who favor diversity might advantage applicants with a migration background.

7.4 Assessing Antidiscrimination Policies

The complex schemes of monitoring and reporting attached to antidiscrimination laws and policies clearly run the risk of only being an attractive but purposeless platform if the operators do not fully commit to the program. Supervising the achievement of programs is, therefore, an inseparable element contributing to their efficiency. In most cases, compliance with monitoring is not guaranteed by sanctions or penalties, and participation in reporting may be far from effective.

In the Netherlands, the assessment of monitoring provided for by the 1994 Wet bevordering evenredige arbeidskansen voor allochtonen (Act on the Promotion of Proportional Labor market Participation of Allochthones; Wet BEAA) demonstrates that only 14% of employers fulfill all of the legal provisions, including the submission of a report on the situation of minorities within the company (Guiraudon et al. 2005). Less than 60% of these had applied for the obligatory registration of the ethnic origin of employees. The Act for Stimulation of Labour Market Participation, which replaced the Wet BEAA in 1998, clearly improved the level of participation, however: In 2001, 70% of employers prepared an annual report detailing the level of representation of ethnic minorities within their company and the measures taken to improve this over the following year. However, while the objectives set representation at 10%, the results reached their ceiling at 8.5%. Although employers with more than 35 people staff were legally obliged to register ethnicity and to submit reports every year, they could also refuse to comply without having to motivate their refusal. The decision to discontinue the SAMEN law in 2003 was partly justified by the lack of participation of employers in the scheme (Guiraudon et al. 2005).

In the UK, the assessment of equality policies is incorporated into the design of the equality programs themselves. Under the Race Relation Act of 2000 (amended), the duties are stricter for public authorities than for private employers. A 1998 survey on the working conditions within companies (Workplace Employee Relations Survey, WERS), which was analyzed in 2003, showed that equality programs are applied within two thirds of companies, 97% of public companies and 57% from the private sector. The programs are implemented more often in companies that have a higher representation of “minorities” (women, ethnic minorities, and disabled people). Among the various actions provided for by the equality programs, the monitoring of employees’ ethnic and racial origin is only carried out by 30% of companies. This disappointing level of monitoring also applies to companies from the public sector, where only 48% of companies have implemented it.

A review by Dex and Purdam (2005) did not find significant improvements after the amendment of the Race Relation Act in UK in 2000: the Commission for Racial Equality found in 2003 that just over a third of organizations were responding to the duties, though most of the public organizations had produced a race equality scheme or policy. In the private sector, a 2003 survey with 500 UK directors identified similar gaps between policies aiming at promoting equal opportunities and the implementation of monitoring system: only 38% of organizations had collected information on the number of employees by ethnic group, and 22% got this information by job positions. In their review of the monitoring practices of ten employers in UK, Dex and Purdam (2005) revealed that although all the employers were collecting data for equal opportunities monitoring purposes, only a few were able to compile these data in tables with standardized categories matching the codes of practice of the Commission for Racial Equality, and hardly any of them were analyzing the data produced (Dex and Purdam 2005, 16–18).

Beyond the assessments of a system’s performance, which is an important condition in assessing its results, a key question remains unanswered: Do the schemes succeed in reducing the consequences of discrimination, easing prejudice, and improving the position of the protected groups? Few programs provide appraisals linking the implementation of initiatives with the improvement of the situation of the protected groups. The Employment Equity Act Annual Reports in Canada, however, are notable exceptions as they provide this type of appraisal. A representation index by group is calculated for each company and business sector. Its variation provides an indication of the impact of the programs. In 2010, the representation of aboriginals, women, and visible minorities had improved, both quantitatively and qualitatively. On the other hand, this remained poor for disabled people. The representation index (the rate of availability relating to the size of a group within the labor force) is established at 95.9 for women, 80.7 for natives and 77.5 for visible minorities but only 46.9 for disabled people.

In the US, a great deal of research has been conducted to assess the impact of affirmative action on employment and education for minorities and women. Holzer and Neumark (2000) demonstrate that the organizations that have adopted the affirmative action programs have seen a clear improvement in the representation of minorities and women in relation to those who did not. However, women, and especially white women, have benefited more from these policies than racial minorities. These findings have been renewed by the evaluation of the outcomes of diversity programs conducted by Dobbin and Kalev (2016). In an assessment of the employment practices and workforce reviews of more than 800 companies in the US from 1971 to 2002, they conclude that mandatory diversity training was producing poor return while programs strengthening managerial responsibility and accountability with respect to equality tended to be particularly effective.

7.5 Conclusion

This chapter has reviewed how policies can address discrimination, with the different frames and tools that have been adopted. The first stage of these policies is to raise awareness and disseminate concepts and definitions of discrimination in legal action. The second and more effective stage aims at monitoring decision-making processes and selection practices to promote equal treatment beyond formal principles. Proactive policies can be called positive action or affirmative action: in all cases, they rely on the existence of statistics broken down by ethnicity, race, or equivalent characteristics to uncover unfair treatment and disadvantage faced by minorities. The lack of such statistics in schools, workplaces, housing, or health systems makes it complicated, if not impossible, to implement most of the schemes of positive action policies. This explains why most European countries fail to develop effective policies against ethnic and racial discrimination, in stark contrast with gender equality programs.

Because antidiscrimination policies address structural inequalities rooted in historical systems of domination, it would be very optimistic to think that they could redress wrongs done by long established and renewed prejudices. For this reason, they have to be judged in the long run. Not only do they need time to effectively tackle discrimination, but their legitimacy is always fragile. If public opinion accepts the implementation of policies and actions targeting minorities when responsibilities of the state are obvious, such support declines dramatically when blatant racism and racial gaps tend to diminish. Opposition to race-based affirmative action or positive action has increased in countries that have pioneered such policies, such as the US. This reminds us that fighting discrimination is not a zero-sum game: when losers improve their position, former winners tend to regret their privileges.