Discrimination profoundly impacts individuals’ lives; restricting economic opportunities, increasing housing and financial insecurity, threatening personal safety, limiting access to cultural, social, and civic activities, and contributing to poor physical and mental health. It also weakens national economic potential and social cohesion. This chapter builds on novel OECD research examining the adverse outcomes linked to discrimination. It finds that, after accounting for self-reported experiences of discrimination, individuals from racialised communities, persons with disabilities, LGBTI people, and religious minorities have outcomes comparable to the general population. This suggests that discrimination itself – rather than group membership – drives disparities in outcomes. These findings reinforce the need for policies to combat discrimination, particularly as evidence in this chapter indicates that strong anti-discrimination laws and inclusion policies are associated with greater social acceptance of minority groups.
Combatting Discrimination in the European Union
Abstract
As highlighted in the previous chapter, discrimination remains a common phenomenon experienced by people from racialised communities, persons with disabilities, lesbian, gay, bisexual, transgender and intersex (LGBTI)1 people and religious minorities. Across the European Union (EU), more than half of people who identify as part of these groups report experiencing discrimination in the past 12 months (Chapter 2). This high prevalence is of concern, not only because it undermines individuals’ dignity and equal opportunities, but also because of its far-reaching consequences. Over the past four decades, research has consistently shown that discrimination negatively affects employment, income, educational attainment, political participation and health, while also generating broader economic and social costs (Pager and Shepherd, 2008[1]; OECD, 2024[2]; 2025[3]). However, most of this evidence has been collected in the United States.
In Europe, the European Union Agency for Fundamental Rights (FRA) regularly surveys groups at risk of discrimination to examine the contexts in which discrimination occurs, and the effects of harassment and hate‑motivated violence on people’s lives. For instance, FRA studies (2023[4]; 2019[5]; 2022[6]; 2024[7]; 2024[8]) indicate that Muslim people, Jewish people, LGBTI people and people of African descent who experience hate‑motivated violence or harassment commonly suffer psychological harm, feel unsafe in public places and are afraid to seek medical attention after a physical attack. While these studies reveal how people react to a discriminatory or hate-filled violent experience, they offer limited information on how the lives of people exposed to discrimination compare with those of the general population. This is largely because FRA’s surveys focus primarily on at-risk groups, rather than on representative samples of the population at large.
Recent OECD research addressed this gap by providing new evidence from general population surveys on how the lives of people who self-report experiences of discrimination in the EU compare to those who do not disclose experiences of discrimination (Hardy and Schraepen, 2024[9]). The analysis examined how the experience of discrimination among people from racialised communities, persons with disabilities, LGBTI people and religious minorities differs from the experiences of discrimination reported by people who do not identify as part of an at-risk group. Once the experience of discrimination was factored into the analysis, there were negligible differences in outcomes between groups at risk of discrimination and the general population – indicating that disparities in outcomes between groups are likely driven by the experience of discrimination (Box 3.1).
Building on this OECD research, this chapter explores less-studied dimensions of life that may be shaped by discrimination. In addition to the economic, mental health, civic engagement and public safety outcomes previously analysed, Section 3.1 examines people’s perceptions of the effects of discrimination, how cultural participation, people’s social lives and physical health are associated with the experience of discrimination. Section 3.2 then turns to the economic and social costs of discrimination, highlighting the role strong anti-discrimination laws and policies can play in fostering social acceptance of groups at risk of discrimination. Indeed, OECD EU countries that are more accepting of minorities also have strong anti‑discrimination laws and inclusion policies, as revealed by the OECD Anti‑Discrimination Questionnaire, completed by 21 out of 22 OECD EU countries. This questionnaire canvasses laws and policies that are aimed at fighting discrimination and promoting equality and inclusion – uncovering gaps in OECD EU countries’ initiatives, as well as good practices, which are discussed in Chapters 4 and 5 (see Annex 3.A for more information about the OECD Anti‑Discrimination Questionnaire).
Discrimination can affect nearly every aspect of people’s lives. It limits access to employment, education, housing, health care, social trust, a sense of safety and willingness to engage in civic and social life. The experience of discrimination in one area of life can trigger negative effects in other areas of life, for instance being discriminated while searching for a job can affect a person’s mental health and lead to economic insecurity, which in turn can affect their ability to secure housing or care for their children.
This section presents evidence on the lives of people who experience discrimination and how their economic, social, civic and health outcomes compare to people who do not self-report discrimination. Drawing on multiple survey sources – including the Opportunities Module of the 2022 Risks that Matter Survey, the 2023 Discrimination in the EU Eurobarometer and various waves of the AXA Mind Health Survey – this analysis offers a multidimensional view of how discrimination affects well-being. These surveys are used in combination because they capture different aspects of life and provide complementary insights.
The analysis is guided by the OECD Well-Being Framework, which encompasses the material conditions that shape people’s economic options (e.g. income and wealth, work and job quality, housing) and factors that affect people’s quality of life, what they know and can do, how they spend their time and the safety, health, and social connectedness of their communities (OECD, 2024[10]). Key well-being indicators are developed for sub‑groups of the population based on individuals’ identification with various groups at risk of discrimination (e.g. ethnicity and skin colour or racial origin, disabilities, sexual orientation and gender identity, and religion). Well-being indicators for people who identify as part of an at-risk group are then compared to the general population – with comparisons specifically made between at-risk groups that report having (or not) experienced discrimination against the general population that also states that they have (not) experienced discrimination, after controlling for a range of other relevant factors such as sex, age, employment status, place of residence and country.
The surveys used in this chapter have the benefit of capturing information on self-reported experiences of discrimination, people’s views on the identity groups they belong to, and indicators of individuals’ material conditions and quality of life. However, there are limitations with using these surveys. As discussed in Chapter 1 and Box 3.1, surveys that ask people to recall experiences can be open to subjective interpretation and may not accord with reality (i.e. discrimination may be observed even though it has not occurred, and vice versa). Notwithstanding this limitation, this chapter primarily uses the Opportunities Module of the 2022 OECD Risks that Matter Survey, which provides a definition of discrimination to increase the consistency with which survey respondents interpret the question.2 Moreover, this section’s analysis is complemented by evidence from other European sources, including experimental studies, which provide a more objective measure of the effects of discrimination (albeit in a restricted number of life domains (Box 3.1)).
In addition, the surveys used in this section are representative of the general population and tend to have small sample sizes (ranging between 500 and 2 000 respondents per country, which translates to a handful of people who identify as part of a group at risk of discrimination, in some cases). While these surveys enable the analysis to draw out the similarities and differences in the experiences of various at-risk groups compared to the general population, small sample sizes make it difficult to conduct a granular analysis of every group at risk of discrimination (and their intersections). As a consequence, this section presents results at the EU-level rather than for each country, and conducts a limited intersectional analysis based on how people’s well-being is shaped by their self-reported experience of discrimination, age, sex and self‑identified at-risk status. The level of granularity of the results also depends on the detail captured by the surveys. In some cases, results cannot be disaggregated by individual at-risk groups, and are instead presented at the level of “at-risk groups” compared to the “non-minority group”.
As discussed in Chapter 1 and Hardy and Schraepen (2024[9]), there are a few broad approaches for estimating the effects of discrimination on individuals – all with advantages and limitations. The first approach is to design experiments (e.g. correspondence and audit studies) to identify the effect of discrimination objectively by developing fictional people who are identical in all ways except for some indication of their at-risk status (e.g. their name or skin colour). Applications for these fictional people are then submitted for jobs or rental properties, and researchers examine whether there are differences in success rates based on names or skin colour or on other markers of minority status. While these approaches can confidently attribute differences in outcomes to discrimination, they can only feasibly be conducted in a few settings, such as job recruitment or rental applications (OECD, 2020[11]; Valfort, forthcoming[12]), and it can be difficult to scale them up for population-level analysis.
Second, statistical approaches can be used to compare the life outcomes of people at risk of discrimination with the rest of the population (e.g. wage levels or health outcomes). Census, administrative and social survey data with rich demographic and economic variables can be used to compute the gap in outcomes between the at-risk group of interest and the general population. After controlling for a range of additional factors that contribute to people’s outcomes (such as age, sex, education and location), the remaining gap is interpreted as indicating the presence of discrimination and other influences that are not directly observed (although because discrimination is not measured, its effects cannot be known with certainty (OECD/European Union, 2015[13]; OECD, 2020[11]; Valfort, forthcoming[12])). In Europe, this type of analysis is limited to the outcomes of persons with disabilities, migrants, women and people at risk of age discrimination, given the paucity of census and representative social survey data on other groups at risk of discrimination – except in countries that collect information on people’s ethnicity, racial origin and sexual orientation (e.g. Ireland and Malta).
Alternatively, self-reported discrimination survey data can be used to examine the effects of discrimination on individuals. Discrimination surveys have the benefit of asking people about their identities and experiences of discrimination, and some include a wide range of information on outcomes of interest such as income, housing, safety and health. This information enables the effect of (self‑reported) discrimination to be observed, after controlling for other explanatory factors. Nevertheless, this approach is not without limitations. It is likely that not all people disclose their identities or experiences of discrimination, while some others may not be aware they experienced discrimination. Further, some people may identify treatment as discriminatory even where this is not the case (OECD/European Union, 2015[13]), although evidence suggests that under‑reporting is more common than over-reporting (Habtegiorgis and Paradies, 2013[14]) (Chapter 1).
Material conditions
Discrimination frequently arises during job searches, in the workplace and when seeking housing (Chapter 2). These experiences can lead to economic insecurity (OECD, 2023[15]), and in turn, reinforce disadvantage by exposing individuals to discrimination based on their socio-economic status.3 As shown in Figure 3.1, people who have experienced discrimination in the previous year are more likely to be at the bottom of the income distribution and to be worried about their financial position and housing affordability over the long term, compared to people who do not report experiencing discrimination. These results hold for all groups – including people who do not identify as part of an at-risk group – although some groups, such as persons with disabilities face a higher likelihood of having poor material conditions across all indicators than other groups.
The heightened risks of poor material conditions for persons with disabilities are consistent with OECD research (2022[16]), which shows persistent disability gaps in employment, unemployment and poverty. The average employment rate for persons with disabilities in Europe is 53%, compared with 77% for others, while employment rates for women with disabilities and young people with disabilities are even lower at 49% and 47%, respectively (European Disability Forum, 2023[17]). While there are many possible reasons why persons with disabilities have lower-than-average employment rates, correspondence studies, which compare the effects of discrimination on various groups, reveal that persons with disabilities face high levels of discrimination in hiring – receiving 40% fewer callbacks than persons without disabilities, according to one study (Lippens, Vermeiren and Baert, 2023[18]). In light of the employment barriers persons with disabilities face, the OECD’s Anti‑Discrimination Questionnaire indicates that over 85% of OECD EU respondents have focused their efforts on boosting employment opportunities for persons with disabilities by developing disabilities employment quotas, employment incentives and programmes to support persons with disabilities in obtaining and keeping a job (Chapter 4).
The nature of employment is also likely affected by discrimination. There is evidence to suggest that people who experience discrimination – particularly people who identify as part of an at-risk group – face higher levels of job insecurity in terms of fear of losing their jobs and being in non-standard work (which includes working without a contract or being on a temporary contract). According to the Opportunities Module of the OECD 2022 Risks that Matter Survey, 50% of people who have experienced discrimination and identify as part of an at-risk group based on their ethnicity or skin colour are concerned about losing their jobs in the next year compared to 40% of people who do not report discrimination but identify as part of these groups. Almost 30% of at-risk women who experience discrimination are in non‑standard work compared to 20% of other women and at-risk men who self-report discrimination.
These results align with qualitative research on migrant women in non-standard work in Europe. The FRA (2018[19]) interviewed 237 female domestic workers about the exploitation they experience. Almost all of the women who participated in the study stated that they experienced bullying, harassment, emotional and/or physical abuse or violence at the hands of their employers or their family members. Many of the women believed that racism and discrimination were the reasons for their exploitation, along with their economic desperation, need to support family members, fear of deportation and uncertainty around their migration status, and lack of power relative to their employer.
People who self-report experiencing discrimination – especially people from racialised communities – are also more likely to be concerned about housing than those who do not disclose discrimination (Figure 3.1). European experimental evidence suggests that discrimination is at play in the difficulties people from racialised communities face in finding housing. For example, an Irish field experiment found that Nigerian applicants were 40% less likely than Polish applicants to receive invitations to view rental accommodation, who were in turn, 15% less likely than Irish applicants to be invited for an inspection (Gusciute, Mühlau and Layte, 2020[20]). Similarly, a recent correspondence test from Belgium found that people of North African descent face discrimination when trying to find an apartment to rent, particularly when looking for apartments in higher socio-economic and less ethnically diverse areas – indicating that people from racialised communities can get “locked into” disadvantaged areas (Ghekiere and Verhaeghe, 2022[21]).
Outcomes for at-risk and non-minority groups by their experience of discrimination in the previous year, EU 17, 2022
Note: Probabilities are derived from logistic models predicting the likelihood of: being in the bottom income quintile; concern about household’s financial situation; concern about housing; concern about being a victim or violence; dissatisfaction with public safety services; or believing that governments do not listen to people like themselves. The income model uses household disposable income and controls for experience of discrimination in the previous year, at-risk identity, sex, age, employment status, occupation, industry, educational level, partner's employment status, place of residence, country, and number of children. In addition to these controls, the models of concern about household’s financial situation and housing concerns control for household income. The model of concerns about being a victim of crime or violence controls for experience of discrimination in the previous year, at-risk identity, age, sex, willingness to pay an addition 2% in taxes for public safety services, household income, education, number of children and country. For the dissatisfaction with public safety services model, the dependent variable is disagreement with the statement “I think that my household and I have/would have access to good quality and affordable public services in the area of public safety (e.g. police)”, and the independent variables are experience of discrimination in the previous year, at-risk identity, age, sex and country. Finally, in the model of perceptions about voice counting, the dependent variable is disagreement with the survey question: “I feel the government incorporates the views of people like me when designing or reforming public benefits and services” and the control variables are experience of discrimination in the previous year, at-risk identity, sex, age, household income, number of children and country. In all models, at-risk identity refers to respondents who consider themselves to belong to an at-risk group based on their ethnicity and skin colour, religion, disabilities and sexual orientation and gender identity (LGBTI). Survey respondents are aged 18-64. The EU 17 average is population weighted and includes the following countries: Austria, Belgium, Denmark, Estonia, Finland, France, Germany, Greece, Ireland, Italy, Latvia, Lithuania, the Netherlands, Poland, Portugal, Slovenia and Spain. All differences in outcomes between those who do and do not experience discrimination are statistically significant at the 10% level except for ethnicity and skin colour minorities at the bottom of the income distribution, ethnic and skin colour minorities who are dissatisfied with public safety services and ethnic and skin colour minorities who do not believe their voice counts.
Source: OECD analysis adapted from Hardy and Schraepen (2024[9]), “The state and effects of discrimination in the European Union”, OECD Papers on Well-being and Inequalities, No. 26, OECD Publishing, Paris, https://doi.org/10.1787/7fd921b9-en, and based on the Opportunities Module of the OECD (2022[22]), Risks that Matter Survey, OECD Publishing, Paris, https://www.oecd.org/en/about/programmes/oecd-risks-that-matter-rtm-survey.html.
Even when people at risk of discrimination find housing, it is more likely to be of poor quality (i.e. lacking access to essential energy services such as heating, hot water, cooling, lighting and energy for appliances (so-called energy poverty)). North African migrants and Syrian migrants experience higher-than-average rates of energy poverty – at 19% and 31%, respectively (compared to an average of 9% across the EU) (European Union Agency for Fundamental Rights, 2024[23]). Twenty-one per cent of Roma people face energy poverty and more than 50% live in situations classified as housing deprivation – with 34% in housing without an indoor shower or bathroom, 33% without an indoor toilet, 19% in housing that is too dark, and 25% living with a leaking roof, damp walls or signs of rot (European Union Agency for Fundamental Rights, 2022[6]). By contrast, the rate of housing deprivation is 17% for the EU (European Union Agency for Fundamental Rights, 2022[6]).
Discrimination may also affect people’s economic opportunities in ways that are difficult to measure. For example, the systemic barriers that at-risk groups face in participating equitably may require them to work harder to achieve similar results to the majority of the population. As such, even if at-risk groups have comparable economic outcomes to the general population, statistics do not show the disproportionate levels of effort they needed to achieve their results. However, experimental studies highlight that at-risk groups need to expend more effort to secure employment. For instance, a field experiment in Sweden revealed that Arab women needed to be more qualified (by having one to three years more experience) than their non-Arabic counterparts to receive the same number of callbacks for advertised jobs (Arai, Bursell and Nekby, 2015[24]). Likewise, religious Muslims face barriers when applying for jobs in France unless they are “outstanding” in terms of graduating from high school with honours, professing a level of mastery of key employment skills, and English fluency (Valfort, 2020[25]).
The effects of discrimination may also extend beyond the immediate outcomes, accumulating over time and across generations to shape individuals’ long-term material conditions. While these effects are difficult to observe in the EU due to limited longitudinal data, American studies indicate that the effects of racial discrimination, for instance, endure. A meta- discrimination analysis of correspondence studies has found that racial discrimination in recruitment has not changed since the late 1980s (Quillian et al., 2017[26]). Reasons for the persistence of discrimination and racial inequality include inadequate access to health care and well‑funded schools for ethnic and racial minorities, housing insecurity and exposure to toxins, and lower household income, wealth and neighbourhood resources (National Academies of Science, Engineering, and Medicine, 2023[27]). These disadvantages begin early: children from racialised communities enter school at a disadvantage compared to their peers – a gap that grows throughout their education and entry into the labour market (Carneiro, Heckman and Masterov, 2005[28]).
Although equivalent longitudinal analysis is scarce in the EU, many OECD EU countries have implemented policy initiatives that recognise the long-term barriers that discrimination creates. More than half of the OECD EU countries that responded to the OECD Anti-Discrimination Questionnaire tailor their educational funding and programmes to students from at-risk groups (Chapter 4). In particular, there are programmes to increase the educational inclusion of Roma students by including students’ families and communities in the school environment. Such initiatives seek to break the intergenerational cycle of exclusion and discrimination by creating fairer opportunities for future generations.
Feelings of safety and participation in cultural, civic and social life
Beyond its material effects, discrimination is associated with a range of negative quality-of-life indicators such as feeling unsafe or perceiving that governments do not listen to people like themselves (Figure 3.1). People who experience discrimination believe they have at least a two‑in‑three chance of being the victim of crime or violence – much higher than people who do not self-report discrimination. Safety concerns are particularly pronounced for people from racialised communities.
In addition to safety concerns, people self‑reporting discrimination are much more likely to be dissatisfied with public safety services, including the police – especially if they are part of a religious minority, LGBT people or not part of an at-risk group. Surprisingly, despite their exposure to racial profiling by the police, racialised people report lower-than-expected dissatisfaction rates. This could reflect a fall in the frequency of police stops of people of African descent, as documented by the European Union Agency for Fundamental Rights between 2016 and 2022 (European Union Agency for Fundamental Rights, 2023[4]). Declining police stops of people of African descent coincides with efforts in many OECD EU countries for police to build trust with racialised communities and eradicate racial bias in policing, including in Austria, Finland, Germany and Luxembourg, which have witnessed significant falls in police stops of people of African descent in the midst of legal and policy changes to prompt more racially inclusive policing practices (Chapter 4). Another possibility is that some racialised individuals have internalised discriminatory treatment as routine and thus do not register dissatisfaction, despite negative experiences.
Discrimination also limits cultural participation and self-expression. People who experience discrimination are two to four times more likely than others to state that they face difficulties in accessing cultural goods, events, places and services (Figure 3.2). Cost and age are most frequently cited as obstacles across the whole population who experience discrimination (Figure 3.2, Panel A), although more than a quarter of people who experience discrimination and identify as part of an at-risk group state that they faced difficulties in participating in cultural activities because they felt uncomfortable or at risk of harassment or because there was a lack of disability accessibility (Figure 3.2, Panel B).
Share of respondents who note barriers to cultural participation and expression, by self-reported experience of discrimination and at-risk status, EU 27, 2023
Note: The survey asked whether respondents face barriers when trying to access cultural goods, events, places and services in their country. Response categories were: Cost, Lack of accessibility for persons with disabilities, Lack of digital knowledge or skills, Feeling too young or too old, Feeling intimidated and at risk of LGBTI-based harassment, Feeling intimidated and at risk of ethnic-origin-based harassment, Feeling intimidated and at risk of skin-color-based harassment, Feeling intimidated and at risk of sex-based harassment; Feeling intimidated and at risk of physical‑appearance-based harassment; Feeling intimidated and at risk of harassment because of being Roma, No, Don’t know. Multiple answers were possible. Panel A shows the results for the entire population, while in Panel B selected barriers are shown for people who identify as part of an at-risk group. Whiskers denote 95% confidence levels. Respondents aged 18 and over were included in the survey and results are shown for the population-weighted average.
Source: OECD calculations based on the European Commission (2023[29]), Discrimination in the European Union, Special Eurobarometer SP535, https://europa.eu/eurobarometer/surveys/detail/2972.
These limitations on cultural participation can contribute to broader civic disengagement. In Europe, cultural activities promote democracy and social cohesion by increasing the likelihood of voting, volunteering and developing positive social attitudes (European Commission, 2023[30]). However, people who experience discrimination are more likely to believe that governments do not listen to them – particularly if they identify as part of a religious minority (Figure 3.1). Other European evidence suggests that people who experience racial, religious or disability discrimination are less likely to vote or get involved in civic activities, even when they express interest in politics and engage in protests and activism against discrimination (Martin, 2017[31]; Mattila and Papageorgiou, 2017[32]; OECD, 2025[33]).
Disengagement from cultural and electoral life may also be illustrative of difficulties maintaining meaningful social relationships. Loneliness is commonly felt by people who identify as part of an at-risk group and experience discrimination. About 35% of at-risk men and women who experience discrimination report feeling lonely “most or all the time”, compared to 20% of at-risk men and women who do not state they have experienced discrimination (Figure 3.3, Panel A).4 This “discrimination effect” is also observed for non‑minority men and women, though to a lesser extent (with 25% of non-minority men and women self‑reporting discrimination feeling lonely ‘most or all the time’) (Figure 3.3, Panel A).
Examining loneliness by age groups presents a more nuanced picture. While the association between discrimination and loneliness is apparent for older individuals and younger non-minorities, loneliness rates are especially elevated among young at-risk groups regardless of whether they report experiencing discrimination (Figure 3.3, Panel B). Indeed, almost 30% of at-risk youth who do not self-report discrimination feel lonely most or all the time – double the rates of non-minority youth and older at-risk people. Young people may be more susceptible to loneliness than older people because they are at a life stage of discovering who they are and moving into new phases of life where they leave established social supports (like going to university or starting a job) (Mental Health Foundation, 2024[34]). Social media may also play a role, although the overall effects are ambiguous. On one hand, social media can help young people connect and build online communities with their peers; but on the other hand, it also exposes them to bullying and hate (Mental Health Foundation, 2024[34]). As noted in Chapter 2, online spaces are one of the most frequently cited areas in which young people experience discrimination, in contrast to older people, which aligns more closely with the interpretation that the dominating effect of social media is that it alienates young people rather than bringing them together.
Strong social support networks do not fully protect people who experience discrimination from feeling lonely. Panels C and D of Figure 3.3 indicate that the chances of lacking strong social supports are unaffected by the experience of discrimination for at-risk groups – even though these groups report higher rates of loneliness. This apparent paradox reflects a broader distinction highlighted in OECD research on social connectedness: while social supports refer to people’s perceptions of the level of support they receive from their inter-personal relationships, loneliness is a subjective experience of being isolated when a person does not wish to be or feeling their needs are not met by their relationships (Mahoney et al., 2024[35]). Similarly, qualitative research in the United Kingdom challenges assumptions that Black and Asian people, as well as other people with ethnic minority backgrounds, cannot be lonely because they often live in large, multi‑generational households (British Red Cross, 2019[36]). In reality, people can live in large households and still feel like they do not have close, meaningful relationships to those around them.
Share of respondents who lack of strong social supports and are lonely, by discrimination experience, at-risk status, sex and age, EU 6, 2022 and 2023
Note: The AXA Mind Health Surveys asked how often over the last four weeks they felt lonely: “All of the time”, “Most of the time”, “Some of the time”, “A little of the time”, “None of the time”, “Don’t know”. Panels A and B show the aggregations of the “All of the time” and “Most of the time” responses. Panels C and D show the responses that strongly disagree and disagree to the statement “I have a great social support network of people I value and trust”. At-risk individuals are those who self-identify as part of an at-risk group based on ethnicity or skin colour, language, disability, sexual orientation or gender identity, religion or belief, migrant status, political opinion or other. Non-minority are those who do not consider themselves to be part of an at-risk group. The EU average is population weighted and based on responses from Belgium, France, Germany, Ireland, Italy and Spain. People aged 18-75 years were surveyed. Whiskers denote 95% confidence intervals.
Source: OECD calculations based on AXA (2024[37]), “Mind your health in the workplace”, 2024 Mind Health Report, https://www-axa-com.cdn.axa-contento-118412.eu/www-axa-com/d41133bc-5fa9-4a5d-b664-316282190d78_axa_mind_health_report_2024.pdf.
These results underscore the importance of addressing loneliness directly, rather than assuming it is a problem that does not exist in communities at risk of discrimination, given perceptions that they are large and close-knit. Loneliness is emerging as a key policy issue, as it is detrimental to physical and mental health, life satisfaction, educational attainment and labour market outcomes, as well as trust in government and support for democratic norms (Mahoney et al., 2024[35]). Qualitative research in the United Kingdom emphasises the need for inclusive services to forge a sense of belonging for Black and Asian people, along with other people with ethnic minority backgrounds, particularly age – and culturally appropriate services that are embedded in the community (British Red Cross, 2019[36]). Chapter 4 discusses efforts in OECD EU countries to design policies, programmes and services in ways that are inclusive of groups at risk of discrimination.
Compounding effects on people’s health
Discrimination has far-reaching implications for physical and mental health. People who experience discrimination are more likely to state that they have poor physical and mental health than those who have not disclosed experiencing discrimination. As shown in Figure 3.4, individuals who self-report discrimination are twice as likely to rate their physical health as poor (Figure 3.4, Panel B), and an almost three times higher likelihood of poor mental health compared to people who do not report experiencing discrimination (Figure 3.4, Panel D). They are also less likely to rate their physical and mental health as “good” compared to people who do not report experiencing discrimination (Figure 3.4, Panels A and C).
These patterns hold for all groups experiencing discrimination – whether based on ethnicity and race, sexual orientation and gender identity, religion, or not part of an at-risk group. The only notable difference across groups is that persons with disabilities are more likely to rate their physical health as bad, which could relate to the nature of their disabilities (Figure 3.4, Panels A and B respectively).
Self-rated physical and mental health status by experience of discrimination in the past 12 months and self-identified at-risk status, EU 6, 2023
Note: The survey asked respondents to evaluate their physical health and their mental health. Respondents could choose from very good, good, average, bad, very bad, and prefer not to say. Panels A and C aggregate “very good” and “good” responses, while Panels B and D aggregate “very bad” and “bad”. The results shown control for age, sex, chronic physical health conditions, mental health conditions, smoking, heavy alcohol consumption, eating habits, exercise, stress levels, income, education attainment, perceived effectiveness of treatment for mental health conditions, loneliness and country. The data represent a population-weighted average across Belgium, France, Germany, Ireland, Italy, and Spain. Whiskers represent 90% confidence intervals.
Source: OECD calculations based on AXA (2024[37]), “Mind your health in the workplace”, 2024 Mind Health Report, https://www-axa-com.cdn.axa-contento-118412.eu/www-axa-com/d41133bc-5fa9-4a5d-b664-316282190d78_axa_mind_health_report_2024.pdf.
Discrimination appears to have a compounding effect on mental health, as people who experience multiple forms of discrimination report higher rates of depression, anxiety and high stress levels (Figure 3.5 ). Experiencing multiple forms of discrimination is associated with poorer mental health outcomes, compared to experiencing a single form of discrimination or none. After controlling for age, sex, income and feelings of loneliness, those experiencing multiple forms of discrimination are 5 percentage points more likely to have depression than people who have experienced a single form of discrimination and 15 percentage points more likely than people who have not been exposed to discrimination (Figure 3.5, Panels A and B). This relationship is consistent across both at-risk and non-minority groups – underscoring that it is the experience of discrimination itself, rather than demographic identity alone, that drives poor health.
Mental health outcomes by experience of discrimination and self-identified at-risk status, EU 6, 2022-2023
Note: This figure shows the share of affirmative responses to questions about people’s current mental health conditions such as depression and anxiety, after controlling for respondents’ age, sex, income, social supports, loneliness and country. Respondents were also asked to rate their stress levels over the past 12 months. Stress levels are measured on a 0-to-10-point scale, with 0 meaning “no stress” and 10 indicating “extremely severe stress”. Self-rated scores above eight are considered high. The share of respondents with high stress levels is shown, after accounting for age, sex, income, social supports, loneliness and country. At-risk individuals are those who self-identify as part of an at-risk group based on ethnicity or skin colour, language, disability, sexual orientation or gender identity, religion or belief, migrant status, political opinion or other. Non-minority are those who do not consider themselves to be part of an at-risk group. The figures shown are the population-weighted average of Belgium, France, Germany, Ireland, Italy and Spain. Whiskers represent 90% confidence intervals.
Source: OECD calculations based on AXA (2024[37]), “Mind your health in the workplace”, 2024 Mind Health Report, https://www-axa-com.cdn.axa-contento-118412.eu/www-axa-com/d41133bc-5fa9-4a5d-b664-316282190d78_axa_mind_health_report_2024.pdf; and AXA (2024[38]), Mind Health Index 2024, https://www-axa-com.cdn.axa-contento-118412.eu/www-axa-com/f5356e90-f204-4848-bac4-aea6b732ff18_axa_mind_health_index_2025.pdf.
These results align with a well-established body of international research. Studies conducted over the past two decades consistently link discrimination with adverse mental health outcomes such as depression, anxiety and psychological distress, as well as physical conditions including hypertension, breast cancer, obesity and substance use (Remes, Mendes and Templeton, 2021[39]; Pascoe and Richman, 2009[40]; Williams et al., 2019[41]; Paradies et al., 2015[42]). The biological mechanism is clear: discrimination activates stress pathways in the brain and body – such as elevated cortisol levels and blood pressure –which can impair immune, metabolic, and cognitive functioning (Berger and Sarnyai, 2015[43]; OECD, 2023[44]). Chronic stress can also weaken self‑control, which can contribute to the uptake of unhealthy behaviours such as smoking and using alcohol and drugs excessively (Pascoe and Richman, 2009[40]) – which is sometimes used as a coping strategy for dealing with the toll of discrimination (Goreis et al., 2020[45]).
Individuals who experience multiple forms of discrimination are also more likely to perceive that one of the most common effects of discrimination is negative emotional states, such as shame, fear and negative self‑esteem (Box 3.2). The FRA has made similar findings in its recent surveys of people of African descent, Muslims and LGBTI people, noting that psychological problems are the most commonly selected response (61% and 55%) among people who report being the victim of severe forms of hatred such as racial violence or (63%) hate-motivated physical or sexual violence (European Union Agency for Fundamental Rights, 2023[4]; 2024[8]; 2024[7]).
Another pathway through which discrimination can affect physical and mental health is via structural inequalities that occur throughout a person’s life. Discrimination and inequalities in opportunities have been called “the causes of the causes” of disease as they shape many health risk factors (Compton and Shim, 2015[46]). For instance, discrimination in employment, education and housing can constrain economic opportunities and lead to people living economically disadvantaged lives where they are at risk of food and energy insecurity and other deprivations that increase the risk of poor health (OECD, 2023[44]; Shim and Compton, 2020[47]; Compton and Shim, 2015[46]).
In addition, the health system may be a site of discrimination, particularly in terms of a lack of access to care and lower quality of care. Research by FRA (2019[5]; 2022[6]) shows that 16% of LGBTI people in EU and 14% of Roma and Travellers experienced discrimination in the previous 12 months in healthcare settings. Similarly, the AXA (2023[48]) Mind Health Survey reveals that people who experience discrimination are less likely to believe the health system provides timely support in treating mental health conditions, and are slightly less likely to feel they know how to access health care if they need it than people who have not experienced discrimination.
Discrimination in health care may result in people not receiving correct diagnoses or adequate treatments or feeling stigmatised. In a recent survey of over 11 000 people from marginalised communities in France, Brazil, Japan, the United Kingdom and the United States revealed that 66% of LGBTI people and 73% of people from racialised communities and persons with disabilities had healthcare experiences that damaged their trust in the system (compared to 56-58% of the general population) (Sanofi, 2022[49]). Maternal mortality for black women is almost four times higher than for white women in the United Kingdom and 2.6 times higher in the United States (House of Commons 2023, 2023[50]; NCHS Health E-Stats, 2023[51]). In the case of the health inequities faced by LGBTI people, a review of studies found that some LGBTI people have been denied medical treatment, and have had healthcare providers make assumptions (and being judgmental) about their sexual practices, and not show respect for trans people’s names or pronouns (Medina-Martinez et al., 2021[52]). Poor treatment discourages people from engaging with the health system unless their medical issues become severe.
Disengagement from the health system, and other essential government services, is a key challenge in the fight against all forms of discrimination, especially systemic discrimination. The next chapter examines what OECD EU countries are doing to tailor their health responses to groups at risk of discrimination in order to overcome the structural barriers people face in fully and freely participating in society (Chapter 4).
How people perceive the effects of discrimination has been understudied relative to other types of perceptions, such as the perceived extent of discrimination and the public perceptions of the effects of inequality on people’s lives and society (OECD, 2023[53]; 2021[54]; 2024[55]). The lack of attention paid to the perceived effects of discrimination is a concern, since this research can give an insight into how closely individuals’ views on discrimination accord with lived reality, and can in turn prove useful when crafting awareness campaigns and other anti-discrimination policy measures.
This chapter provides the first glimpse of people’s perceptions of the effects of discrimination by analysing new questions in the Opportunities Module of the 2022 OECD Risks that Matter Survey on what areas of life people believe to be most affected by discrimination. Results show that a vast majority of people (about 80%) believe that discrimination has negative effects on people’s lives, although people’s perceptions depend on their own experiences of discrimination and their sense of identity. More than half of persons with disabilities, LGBT people and those who do not identify as part of an at‑risk group who experience discrimination believe that negative emotional responses – such as feelings of shame, fear and negative self-esteem – are one of the main effects of discrimination. For members of these groups who do not report discrimination, the corresponding shares range between 35% and 43% (Figure 3.6, Panel A).
By contrast, people from racialised communities and religious minorities who experience discrimination commonly report a lack of economic opportunities as a key effect of discrimination (45% and 41%, respectively) – much larger shares than members of these communities who do not self-report discrimination (27% and 29% respectively; Figure 3.6, Panel A). These results are consistent with more objective measures of the constrained economic opportunities facing people from racialised communities and religious minorities who experience discrimination – particularly among women who are most exposed to working a temporary job or without a contract (discussed above).
While men and women rank negative emotional responses as the most common effect of discrimination, differences between the sexes point to the relative economic precarity women face (Figure 3.6, Panel B). Women in the EU are more likely than men to be single-income-earners and have weaker attachments to the labour market, which puts them at a heightened risk of economic insecurity (OECD, 2023[15]). Women who experience discrimination frequently report a lack of economic opportunity as a key effect of discrimination, with 38% nominating this option compared to 27% of women who do not self-report experiencing discrimination (for men the shares are 31% and 28% respectively). In contrast, men who experience discrimination are more likely than women to state that the main effects of discrimination include constraints on political, religious, social and cultural expression and participation: with 26% of men who experience discrimination selecting this option compared to 19% of women experiencing discrimination.
Perceived effects of discrimination by experience of discrimination, at-risk groups and sex, EU 17, 2022
Note: Survey respondents were asked what the ways are, in their opinion, that discrimination and harassment affect people’s lives in their country. Respondents could choose two options from the following: limiting access to good jobs, business opportunities and education (labelled economic opportunities in the chart), limiting access to housing, health care, justice, and other essential services, constraining political, religious, social and cultural expressions and participation, violence, increasing feelings of shame, fear or low self-esteem, little to no effect or can’t choose. In addition, respondents were asked whether they have experienced discrimination in the past year and whether they identify as part of an at-risk group based on their ethnicity and skin colour, disabilities, sexual orientation and gender identity, religion or none. Survey respondents are aged 18-64. The EU 17 average is population weighted and includes the following countries: Austria, Belgium, Denmark, Estonia, Finland, France, Germany, Greece, Ireland, Italy, Latvia, Lithuania, the Netherlands, Poland, Portugal, Slovenia and Spain.
Source: OECD calculations based on the Opportunities Module of the OECD (2022[22]) Risks that Matter Survey, OECD Publishing, Paris, https://www.oecd.org/en/about/programmes/oecd-risks-that-matter-rtm-survey.html.
Section 3.1 outlined the manifold ways in which discrimination is associated with poor well-being outcomes for the individuals affected, including income insecurity, housing stress, mental ill-health, loneliness, lack of civic engagement, or barriers to cultural participation. While those directly affected bear the brunt of these costs, the effects extend far beyond the individual, impacting the wider economy and society. This section explores the mechanisms through which discrimination can affect even those whose lives it does not directly touch.
Economic costs
In economic terms, discrimination represents a loss of potential output. It hampers people’s ability to realise their equality of opportunity, leading them to work in roles that are not commensurate with their talents or requiring them to work harder than they otherwise would to achieve their goals (thereby diverting their efforts away from other valuable endeavours). Discrimination may also lead to underperformance at work due to the mental health strain, or a suboptimal level of economic development if people at risk of discrimination drop out of the labour market or do not invest in their education, believing their hard work will not be rewarded in terms of higher-paying jobs (OECD, 2020[11]; OECD/European Union, 2015[13]). People belonging to groups at risk of discrimination are more likely than non‑minorities to believe factors outside of a person’s control are key to getting ahead in life and put relatively less stock in the belief that hard work pays off (OECD, 2023[53]).
In addition to the labour market effects, the economic costs of discrimination include both personal and public healthcare expenditure resulting from poor living conditions, mental health strain and hate-motivated violence, for example. Social benefit spending may also be higher in the presence of discrimination for people who cannot find employment suited to their skills and experience (Perlov et al., 2020[56]; van Ballegooij and Moxom, 2018[57]). Similarly, tax revenue may be lower if individuals who face discrimination are unable to earn a sufficient income.
Although the broader economic effects of discrimination are well understood in theoretical terms, quantifying them is challenging since discrimination is rarely observed in data collections (Chapter 1). In the absence of direct measures of discrimination, studies typically estimate the size of the benefit that would accrue to the economy if groups at risk of discrimination had similar employment and earning patterns to the rest of the population. This approach measures the benefit of levelling the playfield (known as closing the gap), but it may overstate the effect of discrimination if other factors explaining differences in outcomes between groups are not considered, such as differences in age distributions of groups, which could affect their years of professional experience and hence earnings. Further, studies rarely account for the investments that countries need to make to achieve parity, for example, increases in funding for educational and training, monitoring and enforcement mechanisms, or the effects of the taxes levied to fund the closing of the gap.
On the other hand, only the costs that can be easily quantifiable are included, which could lead to an underestimation. The primary economic effects that are considered are employment and earnings (and their implications for taxation revenue collections). Some studies include the mental health costs of discrimination, which affect labour force participation and public health expenditure, the costs of hate‑based violence on people’s physical and mental health, and the associated costs on health and justice systems. Notably absent are the costs that stem from the possibilities that never come to fruition – the new ideas, innovation and creativity that could have emerged if people had the opportunity to fully participate in society.
Whether studies over- or under-estimate the costs of discrimination depends largely on the data available. Data collections with rich demographic information on the groups at risk of discrimination and their life outcomes allow a more meaningful analysis of the costs of discrimination (through identifying more costs and alternative explanations of individuals’ outcomes). In Europe, undertaking this type of analysis is stymied by the lack of data on particular groups at risk of discrimination, particularly based on ethnicity and racial origin, LGBTI status or religion (Chapters 1 and 5; Hardy and Schraepen (2024[9])).
However, a handful of studies have estimated the drag that discrimination has on national European economies, at least for some grounds where data are available. In Spain, the direct cost of workforce exclusion of persons with disabilities is estimated to stand at 0.9%-2.4% of GDP, and once the flow-on effects on consumption are included, the costs rise to 1.5%-4% of GDP (Cámara, Martínez and Santero-Sánchez, 2020[58]). This study models various scenarios for the inclusion of persons with disabilities, with one scenario focusing only on closing the labour-force-participation gap between persons with and without disabilities (but not changing the sectors in which people work), and another also closing the education gap, which then contributes to the desegregation of the labour market, as persons with disabilities have the skills to undertake the same work as persons without disabilities.
For LGBT discrimination, the annual costs have been estimated to be 0.21%‑0.43% of GDP in Poland, 0.14%‑0.23% of GDP in Hungary, and 0.63%-1.75% of GDP in Romania (Perlov et al., 2020[56]).5 These costs are calculated based on the assumption that without discrimination LGBT people would experience similar wages and patterns of disease and mental health rates as the general population. In the absence of nationally representative data on the size of the LGBT community in Central and Eastern Europe, these costs assume that LGBT people represent 3% of the total population. However, these cost estimates are similar to recent OECD (2024[2]) research, which uses a rich source of nationally representative microdata to estimate the size of the LGBTI community in the United States. After adjusting for location, demographics, occupation and sector, the report finds that LGBTI people are 7% less likely to be employed and those who are employed receive 7% less than their cisgender, heterosexual peers. If these employment and earnings gaps could be closed by 2050, the United States’ economy could be 2.6% larger (as compared to baseline GDP) – equivalent to an annual increase of 0.1%.
Unlike most cost studies that focus on a single ground of discrimination, some studies estimate the economic costs of discrimination across various grounds. In France, GDP could be 3.6% to 14.1% higher if discrimination on the basis of sex, disability status, residence, and migration status was eliminated (Bon-Maury et al., 2016[59]). The lower end of the range represents the additional GDP from closing the wages gap of these groups at risk of discrimination to the national average, while the upper end of the range closes the gaps in employment status, number of hours worked and the level of education as well.
Across the EU as a whole, the costs of discrimination based on ethnicity and racial origin, sexual orientation, religion, disability, sex and age have been estimated to run into the billions (van Ballegooij and Moxom, 2018[57]). One study quantifies the economic costs of discrimination through a few main channels, although different channels are identified for some groups based on the types of risks they face and data availability.
Ethnicity and race, religion, sexual orientation and age: the costs of discrimination are the lost earnings stemming from the heightened risks of unemployment and poor health. The risks of unemployment and poor health (due to hate-motivated violence) are estimated using logistic regression techniques, which adjust for demographics, education level, income insecurity, and place of residence. These risks are then applied to the number of people who state that they belong to a group at risk of discrimination to calculate the number of people likely to experience these effects. The number of people affected is then converted to a monetary term by multiplying by the average gross wage in the EU to arrive at an estimate of lost economic output (GDP). Lost tax revenue is calculated by assuming a tax rate of 36% on labour (based on the OECD average rate in 2017).
Disability: the costs of discrimination are estimated in terms of the lost earnings resulting from the risk of unemployment and the risk of not completing tertiary education. Weighted likelihoods of unemployment and educational non-completion are estimated based on individuals’ self‑identification of their level of limitations on daily activity. To estimate the size of the affected population, the estimated risks are then applied to the number of people who report being at risk of discrimination because of their disability. Similar to other grounds of discrimination, the number of people affected is then converted to a monetary term by multiplying by the average gross wage in the EU to arrive at an estimate of lost GDP, which is then used to derive the figure of lost tax revenue.
As shown in Table 3.1, the estimated economic costs reflect the size of the groups at risk of discrimination – with those subjected to age discrimination incurring the highest costs by virtue of large swathes of the population being at risk of this type of discrimination.6 Indeed, the estimates are based on the European Social Survey, in which only small fractions of the sample identify as part of a group at risk of discrimination. For example, only 4% of people identify as part of a group at risk of discrimination based on their skin colour or ethnicity, religion, disability, sexual orientation or gender identity in the European Social Survey, compared to 9% in the 2023 Discrimination in the European Union Barometer. As a result of the small samples of minorities and the inability to quantify all relevant costs, the estimates are likely to be conservative.
Note: The annual GDP losses across different grounds of discrimination are not perfectly comparable due to different costs included in the calculation. In particular, the costs of disability discrimination include barriers to education, while health-related costs are quantified for all grounds except disability. Estimates have been adjusted from 2016 to 2022 price levels using Eurostat’s average annual Harmonised Index of Consumer Prices (HICP) (18.6% increase over the period).
Source: OECD analysis adapted from Hardy and Schraepen (2024[9]), “The state and effects of discrimination in the European Union”, OECD Papers on Well-being and Inequalities, No. 26, OECD Publishing, Paris, https://doi.org/10.1787/7fd921b9-en; van Ballegooij and Moxom (2018[57]), Equality and the Fight Against Racism and Xenophobia – Cost of Non-Europe Report; European Parliament, 2018, https://data.europa.eu/doi/10.2861/1791; and Eurostat (2023[60]), Harmonised Indices of Consumer Prices (HICPs) - All Items - Annual Average Indices, https://doi.org/10.2908/TEC00027.
Despite the difficulties in quantifying the costs of discrimination, the extant European literature indicates there is a considerable drag on economic performance related to lost employment, wages and poor health. These, however, are not the only ways in which the effects of discrimination spread beyond those directly affected. Discrimination also undermines social cohesion and the trust that facilitates all interactions in society.
Social cohesion
By definition, discrimination is contrary to social cohesion. As discussed in Section 3.1 and highlighted in Hardy and Schraepen (2024[9]), experiences of discrimination are associated with factors that have the potential to erode social cohesion. This includes feelings of loneliness, and disengagement from society, cultural activities and civic participation. In contrast, socially cohesive societies foster a sense of belonging and fight exclusion and marginalisation to promote the well-being of all including through granting everyone the opportunity of upward social mobility (OECD (2011[61]; 2011[62]), Box 3.3). Public support for social cohesion goals remains strong across Europe, with surveys showing that qualities like “tolerance and respect for others” and a “sense of responsibility” are considered more important for children than individualistic traits like “hard work” (OECD, 2011[61]).
Empirical studies have demonstrated that social cohesion contributes to economic growth. Notably, intergroup cohesion – characterised by the absence of ethnic, linguistic, or religious conflicts – has been shown to have a positive and substantial impact on GDP per capita (Pervaiz and Chaudhary, 2015[63]). Societies that embrace multicultural and liberal values often exhibit higher levels of trust and solidarity, which are associated with economic development (Breidahl, Holtug and Kongshøj, 2018[64]).
Conversely, undermining equality of opportunity and meritocracy as the path towards upwards social mobility creates resentment and disillusionment among those who unfairly miss out. This can give rise to social discontent, and, in some cases, the desire to upend the social hierarchy through more drastic means, including revolution (Hamilton and Hamilton, 2024[65]). Unequal treatment of people can sow mistrust of those who are different, and cause people who are subjected to discrimination to feel like they do not belong in their country or place of residence, which can thwart efforts to cooperate economically (by inhibiting business ventures and lending) and undermine support for governments, public investment and trust in institutions (van Ballegooij and Moxom, 2018[57]; Easterly, Ritzen and Woolcock, 2006[66]; Ekici and Yucel, 2015[67]).
One way to demonstrate the link between social cohesion and economic development is to measure the correlation between the size of the economy and the general population’s support of at-risk groups (using the social acceptance index described above). In the EU, countries with higher GDP per capita are more accepting of people from at-risk groups, although it is unclear whether higher economic development leads to more social cohesion or vice versa (Figure 3.7). Indeed, acceptance of minorities is strongly positively correlated with GDP per capita in EU countries, with an R-squared of 0.7.
Real GDP per capita (2022) and acceptance of minorities (2023) by country
Note: The acceptance of minorities index is constructed by averaging the acceptance levels for: how comfortable one is to have a minority in the highest elected political position, at work with whom they are in daily contact, as a partner of one of their children in a romantic relationship, and to which extent one thinks school lessons and materials should include information about diversity. Levels are aggregated for persons with disabilities, racialised people, Roma people, religious minorities, and LGBTI people. The EU 27 average is population-weighted. Source: OECD analysis adapted from Hardy and Schraepen (2024[9]), “The state and effects of discrimination in the European Union”, OECD Papers on Well-being and Inequalities, No. 26, OECD Publishing, Paris, https://doi.org/10.1787/7fd921b9-en; European Commission (2023[29]), Discrimination in the European Union, Special Eurobarometer SP535, https://europa.eu/eurobarometer/surveys/detail/2972; and Eurostat (2024[68]), Real GDP per capita, https://doi.org/10.2908/SDG_08_10.
To explore the relationship between social cohesion and discrimination, this section presents analysis of social acceptance of minorities and several indicators of social cohesion. These include societies’ willingness to undertake anti-discrimination action, perceptions that governments are not doing anything to promote diversity, the diversity of social groups and views that “my voice counts in my country” (Figure 3.8). The social acceptance of minorities index is based on the level of public support – on a 0 to 10 scale, where 0 stands for “not at all comfortable” and 10 stands for “totally comfortable” – people have for: being in close working relationships with people from minority groups; people from minority groups being in political leadership positions; their children being in romantic relationships with children from minority groups; information on diversity included in school lessons. An index is created for each at-risk group individually – racialised communities, persons with disabilities, LGBTI people, Roma and religious minorities – and then averaged to obtain the overall social acceptance of minorities index.
Despite rising self-reported rates of discrimination in most EU countries (as shown in Chapter 2 and in Hardy and Schraepen (2024[9])), social acceptance of minority groups is generally high – and rising – across the EU (Figure 3.9). Between 2019 and 2023, the level of acceptance of all minority groups rose in most EU countries, with the largest increases seen in Czechia, Finland, Greece, Italy, Latvia, and Lithuania. EU countries are gradually converging towards higher levels of acceptance, particularly for LGBTI people. However, acceptance remains relatively low for LGBTI and Roma people in a few countries, which pulls down the average level. Bulgaria, Romania, Hungary and the Slovak Republic record low levels of LGBTI acceptance, while for Roma acceptance, Bulgaria’s score is relatively lower than other EU countries. Further, acceptance fell between 2019 and 2023 in Romania (mainly due to a decrease in the acceptance of people with an ethnicity different than the majority) and in Poland (lower support for persons with disabilities and a religion different than the majority).
Note: The acceptance of minorities index is constructed by averaging the acceptance levels for: how comfortable one is to have a minority in the highest elected political position, at work with whom they are in daily contact, as a partner of one of their children in a romantic relationship, and to which extent one thinks school lessons and materials should include information about diversity. An index is created for each group at risk of discrimination: persons with disabilities, racialised people, Roma people, religious minorities, and LGBTI people. The acceptance of minorities index is the simple average of these individual indexes. In Panel A, anti-discrimination engagement reflects the share of the population that has done one of the following actions: shared online content about discriminatory incidents, publicly defended someone who was the victim of discrimination, joined an association or campaign that defends people against discrimination and/or publicly raised the issue of discrimination in the workplace in the past year (see Chapter 2 and Hardy and Schraepen (2024[9])). In Panel B, the share of the population with diverse friends or acquaintances is based on responses to the question “Do you have friends or acquaintances that are…?” The share is the unweighted average of whether the respondent has a friend or acquaintance for each of the following five groups: persons with disabilities, ethnicity or skin colour different from the respondent, Roma people, religion different from the respondent, and LGBTI people. Panel C shows the share of people who indicate that their voice counts in their country. Panel D shows the share of people that responded “yes” to the question “Do you think enough is being done to promote diversity in the area where you live by your local authority?”.
Source: OECD calculations based on the European Commission (2023[29]), Discrimination in the European Union, Special Eurobarometer SP535, https://europa.eu/eurobarometer/surveys/detail/2972.
Note: The acceptance of minorities index is constructed by averaging the acceptance levels for how comfortable one is to have a minority: in the highest elected political position; at work with whom they are in daily contact; as a partner of one of their children in a romantic relationship; and to which extent one thinks school lessons and materials should include information about diversity. Levels are computed separately for persons with disabilities, racialised people, Roma people, religious minorities, and LGBTI people. Panel F includes the acceptance for all aforementioned minority groups (equally weighted). Cyprus, Luxembourg and Malta are not presented in the chart due to small sample sizes, but they are included in the population weighted European Union average, which is represented as EU 27.
Source: OECD analysis adapted from Hardy and Schraepen (2024[9]), “The state and effects of discrimination in the European Union”, OECD Papers on Well-being and Inequalities, No. 26, OECD Publishing, Paris, https://doi.org/10.1787/7fd921b9-en; and the European Commission (2023[29]), Discrimination in the European Union, Special Eurobarometer SP535, https://europa.eu/eurobarometer/surveys/detail/2972.
As discussed in Chapter 2, the rise in the social acceptance of at-risk groups has occurred during a period in which self-reported discrimination rates have grown in most EU countries. This apparently paradoxical relationship can be explained by the wide range of factors that influence self-reported discrimination rates. Self-reported discrimination rates reflect demographic, social, cultural and political factors, including the diversity of a country’s population and people’s willingness to attribute their treatment to discrimination (Chapter 2 and Hardy and Schraepen (2024[9])). In countries with greater social acceptance, people may be more likely to disclose experiences of discrimination, partly because they feel more comfortable sharing and may be more knowledgeable about what constitutes discrimination, as accepting societies tend to disseminate more information about discrimination (Chapters 2 and 4, and Hardy and Schraepen (2024[9])).
Moreover, rising social acceptance coincides with concerted EU and national-level legislative and policy efforts to fight discrimination, as symbolised by the introduction of the Union of Equality in 2019, and the expansion of rights in some EU countries (such as the recent recognition of same‑sex marriage in Estonia, Greece and Slovenia) (Chapter 4). The introduction of inclusion policies and laws is linked to increase the level of social acceptance across OECD and EU countries, for example through workplace policies (OECD, 2020[69]), the introduction of legal protections for LGBTI people (OECD, 2020[11]) and migrant integration policies (Solano and Huddleston, 2020[70]).
Findings from the OECD Anti-Discrimination Questionnaire support these observations. OECD EU countries with stronger laws, policies and mainstreaming initiatives to fight discrimination and promote equality and inclusion are more accepting of groups at risk of discrimination (Figure 3.10). In general, the OECD EU countries that responded to the OECD Anti-Discrimination Questionnaire7 have focused their efforts to combat ethnic and racial, and disability discrimination – reflecting the relatively high levels of social acceptance of these groups, their exposure and vulnerability to discrimination (in terms of the extent of discrimination they experience and the severity of its effects), and the strength of EU protections for these grounds (Figure 3.10, Panels A and B, Chapter 4). Nevertheless, some countries including Denmark, France, Finland, the Netherlands, Luxembourg, Spain and Sweden, have pursued equality policies broadly across these grounds, which corresponds to high levels of social acceptance for people from racialised communities, persons with disabilities, LGBTI people, and religious minorities (Figure 3.10, Panels A, B, C and D respectively).
Laws and policies that prohibit discrimination and promote equality and inclusion shape societal values, norms and expectations, which can help to combat the prejudice, stigma and stereotypes that drive discrimination (Office of the United Nations High Commissioner for Human Rights, 2023[71]). The developers of the Migrant Integration Policy Index (Solano and Huddleston, 2020[70]) note that “integration policies are one of the strongest factors shaping the public’s willingness to accept and interact with immigrants”, while evidence shows that legal changes to promote LGBTI equality improve people’s attitudes towards LGBTI people (OECD, 2020[11]). People perceive legal changes as a reflection of what is socially acceptable, and many are willing to adopt these norms (Tankard and Paluck, 2017[72]; OECD, 2020[11]). This is exemplified by the rapid increase in acceptance of homosexuality in European countries after the passage of same-sex marriage laws (Aksoy, 2020[73]).
Correlation between social acceptance of various minority groups and policy, law and mainstreaming indices derived from responses to the OECD Anti-Discrimination Questionnaire, EU 21
Note: The acceptance of minorities scores are constructed for each at-risk group by averaging the acceptance levels for: how comfortable one is to have a minority in the highest elected political position, at work with whom they are in daily contact, as a partner of one of their children in a romantic relationship, and to which extent one thinks school lessons and materials should include information about diversity (see Hardy and Schraepen (2024[9]) for further details). The “initiative indices” capture information on OECD EU countries’ laws, policies and mainstreaming efforts to fight discrimination and promote equality and inclusion, based on responses to the OECD Anti-Discrimination Questionnaire. The construction of these indices is described in Annex 3.A. The interpretation of the initiative indices is as follows: a score of 1 means that the country has comprehensive anti-discrimination laws (prohibiting a broad range of forms of discrimination across key areas of life, permitting positive actions and requiring reasonable accommodations for persons with disabilities), as well as a suite of policies and programmes that are tailored to the needs of at-risk groups in employment, education and training, health, justice, public accessibility and awareness campaigns. Finally, a score of 1 also requires a country to have robust equality and inclusion mainstreaming practices in relation to policy coordination mechanisms, stakeholder engagement processes, staff training, policy and programme evaluation and data collections. In terms of country coverage, while Belgium is included in the figure, it is excluded from the lines of best fit, as it is not directly comparable with the other OECD EU countries that responded to the OECD Anti-Discrimination Questionnaire. Belgium’s response to the OECD Anti-Discrimination Questionnaire comprised national and regional policies and mainstreaming efforts, which differs from the national approach of all other questionnaire respondents. Belgium is included in this figure to show the scope of its anti-discrimination activities across at-risk groups, which can highlight whether Belgium has comprehensively rolled out initiatives for all at-risk groups or whether efforts have focused on particular groups. The analysis is based on national government responses from Austria, Czechia, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Hungary, Latvia, Lithuania, Luxembourg, the Netherlands, Poland, Portugal, Slovak Republic, Slovenia, Spain and Sweden, and national and regional responses from Belgium.
Source: OECD calculations based on European Commission (2023[29]), Discrimination in the European Union, Special Eurobarometer SP535, https://europa.eu/eurobarometer/surveys/detail/2972 and responses to the OECD Anti-Discrimination Questionnaire.
Conclusion
This chapter has shown that discrimination is associated with negative well-being outcomes, such as income and housing insecurity, weaker social, cultural and civic participation, a lack of a sense of safety and poor health. After accounting for self-reported experiences of discrimination, the analysis revealed that individuals from racialised communities, persons with disabilities, LGBTI people, and religious minorities have outcomes comparable to the general population. This suggests that it is discrimination itself – rather than group membership – that drives disparities in outcomes.
In addition, the chapter has examined how the effects of discrimination extend beyond those directly affected. Discrimination constrains economic potential by frustrating people’s efforts to realise their equality of opportunity. This leads them to work in roles that are not commensurate to their talents, or requires them to work harder than they otherwise would to achieve their goals (thereby diverting their efforts away from other valuable endeavours). Discrimination may also lead to underperformance at work due to the mental health strain, or a less-than-optimal level of economic development if people at risk of discrimination drop out of the labour market or do not invest in their education, believing their hard work will not be rewarded in terms of higher-paying jobs.
Moreover, the chapter has shown how discrimination can erode social cohesion by marginalising at-risk groups – denying them opportunities to participate in the economy and society. Discrimination can cause people to feel like they do not belong, which can thwart economic cooperation (by inhibiting business ventures and lending) and undermine support for governments, public investment and trust in institutions. It can also compromise people’s participation in social and cultural events, which may strain community development and civic engagement.
Finally, the chapter has presented evidence that EU countries with strong anti-discrimination protections and inclusive policies are more accepting of minorities. This highlights a potential pathway for combatting discrimination and promoting social cohesion, since minority acceptance is a key component of social cohesion. The next chapter delves more deeply into what OECD EU countries are doing to combat discrimination and promote equality and inclusion. It covers anti-discrimination laws and policies in key areas of life such as employment, education and training, health, justice and safety, and access to public spaces. Good examples in each of these areas are raised, which can serve as inspiration to other countries seeking to eliminate discrimination and foster inclusion and social cohesion.
References
[73] Aksoy, C. (2020), “Do laws shape attitudes? Evidence from same-sex relationship recognition policies in Europe”, European Economic Review, Vol. 124, https://doi.org/10.1016/j.euroecorev.2020.103399.
[24] Arai, M., M. Bursell and L. Nekby (2015), “The reverse gender gap in ethnic discrimination: Employer stereotypes of men and women with Arabic names”, International Migration Review, Vol. 50, pp. 385-412, https://doi.org/10.1111/imre.12170.
[37] AXA (2024), Mind your health in the workplace, https://www-axa-com.cdn.axa-contento-118412.eu/www-axa-com/d41133bc-5fa9-4a5d-b664-316282190d78_axa_mind_health_report_2024.pdf.
[38] AXA (2024), “Mind your health in the workplace”, 2024 Mind Health Report, https://www-axa-com.cdn.axa-contento-118412.eu/www-axa-com/d41133bc-5fa9-4a5d-b664-316282190d78_axa_mind_health_report_2024.pdf.
[48] AXA Group (2023), Towards a New Understanding: How We Strengthen Mind Health and Wellbeing at Home, at Work and Online, https://www.axa.com/en/about-us/mind-health-report.
[43] Berger, M. and Z. Sarnyai (2015), ““More than skin deep”: Stress neurobiology and mental health consequences of racial discrimination”, Stress, Vol. 18/1, pp. 1-10, https://doi.org/10.3109/10253890.2014.989204.
[59] Bon-Maury, G. et al. (2016), Le Coût Economique des Discriminations, https://www.strategie.gouv.fr/publications/cout-economique-discriminations.
[64] Breidahl, K., N. Holtug and K. Kongshøj (2018), “Do shared values promote social cohesion? If so, which? Evidence from Denmark”, European Political Science Review, Vol. 10/1, pp. 97-118, https://doi.org/10.1017/S1755773916000266.
[36] British Red Cross (2019), Barriers to Belonging An Exploration of Loneliness Among People from Black, Asian and Minority Ethnic Backgrounds, https://www.redcross.org.uk/about-us/what-we-do/we-speak-up-for-change/barriers-to-belonging.
[58] Cámara, A., I. Martínez and R. Santero-Sánchez (2020), “Macroeconomic cost of excluding persons with disabilities from the workforce in Spain”, IZA Journal of Labor Policy, Vol. 10/15, https://doi.org/10.2478/izajolp-2020-0015.
[28] Carneiro, P., J. Heckman and D. Masterov (2005), “Labor market discrimination and racial differences in premarket factors”, Journal of Law and Economics, Vol. 48/1, pp. 1-39.
[46] Compton, M. and R. Shim (2015), “The social determinants of mental health”, Focus: The Journal of Lifelong Learning in Psychiatry, Vol. 13/4, pp. 419-425.
[66] Easterly, W., J. Ritzen and M. Woolcock (2006), “Social cohesion, institutions, and growth”, Economics & Politics, Vol. 18/2, pp. 103-120, https://doi.org/10.1111/j.1468-0343.2006.00165.x.
[67] Ekici, T. and D. Yucel (2015), “What determines religious and racial prejudice in Europe? The effects of religiosity and trust”, Social Indicators Research, Vol. 122, pp. 105-133, https://doi.org/10.1007/s11205-014-0674-y.
[30] European Commission (2023), Culture and Democracy: The Evidence: How Citizens’ Participation in Cultural Activities Enhances Civic Engagement, Democracy and Social Cohesion, https://doi.org/10.2766/39199.
[29] European Commission (2023), Discrimination in the European Union, Special Eurobarometer SP535, https://europa.eu/eurobarometer/surveys/detail/2972.
[75] European Commission Against Racism and Intolerance (2021), ECRI’s Opinion on the Concept of “Racialisation” - Adopted at ECRI’s 87th Plenary Meeting on 8 December 2021, https://rm.coe.int/ecri-opinion-on-the-concept-of-racialisation/1680a4dcc2.
[17] European Disability Forum (2023), The Right to Work: The Employment Situation of Persons with Disabilities in Europe, European Human Rights Report, Issue 7, https://www.edf-feph.org/content/uploads/2023/05/hr7_2023_press-accessible.pdf.
[74] European Network Against Racism (2019), Lessons for Effective National Anti-Racism Policies: National Action Plans Against Racism, https://www.enar-eu.org/wp-content/uploads/napar_lessons_for_effective_national_anti-racism_policies_final.pdf.
[7] European Union Agency for Fundamental Rights (2024), Being Muslim in the EU: Experiences of Muslims, https://fra.europa.eu/sites/default/files/fra_uploads/fra-2024-being-muslim-in-the-eu_en.pdf.
[23] European Union Agency for Fundamental Rights (2024), Fundamental Rights Report - 2024, https://fra.europa.eu/sites/default/files/fra_uploads/fra-2024-fundamental-rights-report-2024_en.pdf.
[8] European Union Agency for Fundamental Rights (2024), LGBTIQ Equality at a Crossroads: Progress and Challenges, https://fra.europa.eu/sites/default/files/fra_uploads/fra-2024-lgbtiq-equality_en.pdf.
[4] European Union Agency for Fundamental Rights (2023), Being Black in the EU: Experiences of People of African Descent, https://fra.europa.eu/sites/default/files/fra_uploads/fra-2023-being-black_in_the_eu_en.pdf.
[6] European Union Agency for Fundamental Rights (2022), Roma in 10 European Countries, https://fra.europa.eu/sites/default/files/fra_uploads/fra-2022-roma-survey-2021-main-results2_en.pdf.
[5] European Union Agency for Fundamental Rights (2019), A Long Way to Go for LGBTI Equality, https://fra.europa.eu/sites/default/files/fra_uploads/fra-2020-lgbti-equality-1_en.pdf.
[19] European Union Agency for Fundamental Rights (2018), Out of Sight: Migrant Women Exploited in Domestic Work, https://fra.europa.eu/sites/default/files/fra_uploads/fra-2018-migrant-women-labour-exploitation-domestic-work_en.pdf.
[68] Eurostat (2024), Real GDP per capita, https://doi.org/10.2908/SDG_08_10.
[60] Eurostat (2023), Harmonised Indices of Consumer Prices (HICPs) - all items - annual average indices, https://doi.org/10.2908/TEC00027.
[21] Ghekiere, A. and P. Verhaeghe (2022), “How does ethnic discrimination on the housing market differ across neighborhoods and real estate agencies?”, Journal of Housing Economics, Vol. 55, https://doi.org/10.1016/j.jhe.2021.101820.
[45] Goreis, A. et al. (2020), “What mediates the relationship between ethnic discrimination and stress? Coping strategies and perceived social support of Russian immigrants in Germany”, Frontiers in Psychiatry, Vol. 11, pp. 1-14, https://doi.org/10.3389/fpsyt.2020.557148.
[20] Gusciute, E., P. Mühlau and R. Layte (2020), “Discrimination in the rental housing market: A field experiment in Ireland”, Journal of Ethnic and Migration Studies, Vol. 48/3, pp. 613-634, https://doi.org/10.1080/1369183X.2020.1813017.
[14] Habtegiorgis, A. and Y. Paradies (2013), “Utilising self-report data to measure racial discrimination in the labour market”, The Australian Journal or Labour Economics, Vol. 16/1, pp. 5-41, https://ajle.org/index.php/ajle_home/article/view/131 (accessed on 22 May 2025).
[65] Hamilton, C. and M. Hamilton (2024), The Privileged Few, Polity Press, United Kingdom.
[9] Hardy, S. and T. Schraepen (2024), “The state and effects of discrimination in the European Union”, OECD Papers on Well-being and Inequalities, No. 26, OECD Publishing, Paris, https://doi.org/10.1787/7fd921b9-en.
[50] House of Commons 2023 (2023), Black Maternal Health, Third Report of Session, https://committees.parliament.uk/publications/38989/documents/191706/default/.
[18] Lippens, L., S. Vermeiren and S. Baert (2023), “The state of hiring discrimination: A meta-analysis of (almost) all recent correspondence experiments”, European Economic Review, Vol. 151, https://doi.org/10.1016/j.euroecorev.2022.104315.
[35] Mahoney, J. et al. (2024), “Measuring social connectedness in OECD countries: A scoping review”, OECD Papers on Well-being and Inequalities, No. 28, OECD Publishing, Paris, https://doi.org/10.1787/f758bd20-en.
[31] Martin, N. (2017), “Are British Muslims alienated from mainstream politics by Islamophobia and British foreign policy?”, Ethnicities, Vol. 17/3, pp. 350-370.
[32] Mattila, M. and A. Papageorgiou (2017), “Disability, perceived discrimination and political participation”, International Political Science Review, Vol. 38/5, pp. 505-519.
[52] Medina-Martinez, J. et al. (2021), “Health inequities in LGBT people and nursing interventions to reduce them: A systematic review”, International Journal of Environmental Research and Public Health, Vol. 18/22, pp. 1-16.
[34] Mental Health Foundation (2024), Loneliness in Young People: Research Briefing, https://www.mentalhealth.org.uk/our-work/public-engagement/unlock-loneliness/loneliness-young-people-research-briefing#:~:text=These%20findings%20fly%20in%20the,new%20job%20away%20from%20home (accessed on 20 December 2024).
[27] National Academies of Science, Engineering, and Medicine (2023), Reducing Intergenerational Poverty, National Academies Press, https://doi.org/10.17226/27058.
[51] NCHS Health E-Stats (2023), Maternal Mortality Rates in the United States, 2021, https://doi.org/10.15620/cdc:124678.
[33] OECD (2025), Challenges to the Political Representation of Immigrants and Their Children in Germany: A Case Study, https://www.oecd.org/en/blogs/2025/03/supporting-opportunities-insights-from-communities/challenges-to-the-political-representation-of-immigrants-and-their-children-in-germany-a-case-study.html (accessed on 9 April 2025).
[3] OECD (2025), Monitoring and Assessing the Impact of National Action Plans Against Racism, OECD Publishing, Paris, https://doi.org/10.1787/13e32676-en.
[55] OECD (2024), “Enhancing opportunities by design: Exploring people’s views of what should be done to fight inequality”, OECD Policy Insights on Well-being, Inclusion and Equal Opportunity, No. 15, OECD Publishing, Paris, https://doi.org/10.1787/f3a168ef-en.
[10] OECD (2024), How’s Life? 2024: Well-being and Resilience in Times of Crisis, OECD Publishing, Paris, https://doi.org/10.1787/90ba854a-en.
[2] OECD (2024), The Economic Case for Greater LGBTI+ Equality in the United States, OECD Publishing, Paris, https://doi.org/10.1787/5adfe3a5-en.
[53] OECD (2023), “Hard work, privilege or luck? Exploring people’s views of what matters most to get ahead in life”, OECD Policy Insights on Well-being, Inclusion and Equal Opportunity, No. 12, OECD Publishing, Paris, https://doi.org/10.1787/e1903ab2-en.
[44] OECD (2023), How to Make Societies Thrive? Coordinating Approaches to Promote Well-being and Mental Health, OECD Publishing, Paris, https://doi.org/10.1787/fc6b9844-en.
[15] OECD (2023), On Shaky Ground? Income Instability and Economic Insecurity in Europe, OECD Publishing, Paris, https://doi.org/10.1787/9bffeba6-en.
[16] OECD (2022), Disability, Work and Inclusion: Mainstreaming in All Policies and Practices, OECD Publishing, Paris, https://doi.org/10.1787/1eaa5e9c-en.
[22] OECD (2022), Risks that Matter Survey, OECD Publishing, Paris, https://www.oecd.org/en/about/programmes/oecd-risks-that-matter-rtm-survey.html.
[54] OECD (2021), Does Inequality Matter? : How People Perceive Economic Disparities and Social Mobility, OECD Publishing, Paris, https://doi.org/10.1787/3023ed40-en.
[69] OECD (2020), All Hands In? Making Diversity Work for All, OECD Publishing, Paris, https://doi.org/10.1787/efb14583-en.
[11] OECD (2020), Over the Rainbow? The Road to LGBTI Inclusion, OECD Publishing, Paris, https://doi.org/10.1787/8d2fd1a8-en.
[61] OECD (2011), Perspectives on Global Development 2012: Social Cohesion in a Shifting World, OECD Publishing, Paris, https://doi.org/10.1787/persp_glob_dev-2012-en.
[62] OECD (2011), Society at a Glance 2011: OECD Social Indicators, OECD Publishing, Paris, https://doi.org/10.1787/soc_glance-2011-en.
[13] OECD/European Union (2015), Indicators of Immigrant Integration 2015: Settling In, OECD Publishing, Paris/European Union, Brussels, https://doi.org/10.1787/9789264234024-en.
[71] Office of the United Nations High Commissioner for Human Rights (2023), Protecting Minority Rights: Practical Guide for Developing Comprehensive Anti-Discrimination Legislation, https://www.ohchr.org/sites/default/files/documents/publications/2022-11-28/OHCHR_ERT_Protecting_Minority%20Rights_Practical_Guide_web.pdf.
[1] Pager, D. and H. Shepherd (2008), “The sociology of discrimination: Racial discrimination in employment, housing, credit, and consumer markets”, Annual Review of Sociology, Vol. 1/34, pp. 181-209, https://doi.org/10.1146/annurev.soc.33.040406.131740.
[42] Paradies, Y. et al. (2015), “Racism as a determinant of health: A systematic review and meta-analysis”, PLoS ONE, Vol. 10/9, https://doi.org/10.1371/journal.pone.0138511.
[40] Pascoe, E. and L. Richman (2009), “Perceived discrimination and health: A meta-analytic review”, Psychology Bulletin, Vol. 135/4, pp. 531-554, https://doi.org/10.1037/a0016059.
[56] Perlov, G. et al. (2020), The Economic Case for LGBTI+ Inclusion in Central and Eastern Europe (CEE): Hungary, Poland, Romania and Ukraine, https://static1.squarespace.com/static/5bba53a8ab1a62771504d1dd/t/6085c5a9833eae5a86c674c0/1619379629514/The+Economic+Case+for+LGBT%2B+Inclusion+in+CEE+-+ENGLISH.pdf.
[63] Pervaiz, Z. and A. Chaudhary (2015), “Social cohesion and economic growth: An empirical investigation”, Australian Economic Review, Vol. 48/4, pp. 369-381, https://doi.org/10.1111/1467-8462.12128.
[26] Quillian, L. et al. (2017), “Meta-analysis of field experiments shows no change in racial discrimination in hiring over time”, Proceeding of the National Academy of Sciences, Vol. 114/41, pp. 10870-10875, https://doi.org/10.1073/pnas.1706255114.
[39] Remes, O., F. Mendes and P. Templeton (2021), “Biological, psychological, and social determinants of depression: A review of recent literature”, Brain Sciences, Vol. 11/12, p. 1633, https://doi.org/10.3390/brainsci11121633.
[49] Sanofi (2022), A Million Conversations: How We’re Bridging the Healthcare ’Trust Gap’ with Marginalized Communities, https://www.sanofi.com/en/magazine/social-impact/global-poll.
[47] Shim, R. and M. Compton (2020), “The social determinants of mental health: Psychiatrists’ roles in addressing discrimination and food insecurity”, Focus: The Journal of Lifelong Learning in Psychiatry, Vol. 18/1, pp. 25-30, https://doi.org/10.1176/appi.focus.20190035.
[70] Solano, G. and T. Huddleston (2020), Migrant Integration Policy Index (MIPEX) 2020, https://www.mipex.eu/anti-discrimination.
[72] Tankard, M. and E. Paluck (2017), “The effect of a Supreme Court decision regarding gay marriage on social norms and personal attitutudes”, Psychological Science, Vol. 28/9, pp. 1334-1344, https://doi.org/10.1177/0956797617709594.
[25] Valfort, M. (2020), “Anti-Muslim discrimination in France: Evidence from a field experiment”, World Development, Vol. 135, https://doi.org/10.1016/j.worlddev.2020.105022.
[12] Valfort, M. (forthcoming), Strengthening the Evidence base for Racial/Ethnic Disadvantage in OECD Countries, OECD Publishing, Paris.
[57] van Ballegooij, W. and J. Moxom (2018), Equality and the Fight Against Racism and Xenophobia: Cost of Non-Europe Report, https://doi.org/10.2861/1791.
[41] Williams, D. et al. (2019), “Understanding how discrimination can affect health”, Health Services Research, Vol. 54, pp. 1374-1388, https://doi.org/10.1111/1475-6773.13222.
The indices presented in Figure 3.10 are derived from responses to the OECD Anti-Discrimination Questionnaire, which was developed to gather information on laws and policies to promote inclusion and fight discrimination based on a person’s ethnicity or race, disability status, sexual orientation or gender identity, religion and age. Twenty-one OECD EU countries responded to the questionnaire and provided examples of the current, or planned, policies and mainstreaming initiatives (current as of April 2024). To ensure consistency across countries, the OECD supplemented these responses with desktop research. Further details on the law and policy aspects of the questionnaire can be found in Chapter 4, while Chapter 5 focuses on equality and inclusion mainstreaming.
The OECD Anti‑Discrimination Questionnaire comprises three components.
The law component: Guided by the Office of the United Nations High Commissioner for Human Rights (2023[71]), the law component focuses on comprehensive anti‑discrimination legislation that addresses all areas of life regulated by law – including employment, education and training, the health, social services, public safety and justice, online and in the media, the provision of and access to products and housing – while prohibiting various forms of discrimination. The questionnaire covers direct discrimination, indirect discrimination, victimisation, segregation, harassment, hate crimes, hate speech, multiple discrimination, intersectional discrimination, discrimination by association and discrimination by assumption (definitions are provided in Chapter 4). It also includes measures to ensure access to rights, such as reasonable accommodation for persons with disabilities and positive actions to redress disadvantage.
The policy component: In addition to legal prohibitions on discrimination, the questionnaire collects information on policies that promote equality of opportunity and social equality for people at risk of discrimination. It captures policies and programmes designed around the specific needs and circumstances of those at risk of discrimination. This includes ways to: dismantle barriers in education, health and justice services; create incentives to improve economic opportunities; develop campaigns to increase awareness of the nature of discrimination; and design more inclusive societies. To gain a holistic view of key government services, the questionnaire covers early childhood education and care (ECEC), school, tertiary education and transitions to work, community-based primary health services, hospitals, preventative health care, police, and legal assistance, among others.
The mainstreaming component: This component examines efforts to embed non-discrimination and equality throughout government processes (mainstreaming), including cross-departmental coordination, national equality strategies, inclusive stakeholder engagement, equality data collections, policy evaluations, funding for equality initiatives, and training for public officials.
Three separate indices have been developed to enable comparisons of the protections afforded to the groups at risk of discrimination that are examined in this study. For each at-risk group, a law index, a policy index, and a mainstreaming index have been created (Annex Figure 3.A.1 visually represents each index).
The law index reflects the breadth and depth of anti-discrimination laws. It combines information on the number of forms of discrimination that are prohibited, the areas of life covered by anti‑discrimination laws, whether positive actions are permitted and whether there is a requirement to make reasonable accommodations for persons with disabilities. First, a score is given based on the number of forms of discrimination that are prohibited for each protected ground (where a score of 1 is given if all of the following are covered in law: direct and indirect discrimination, harassment, discrimination by association, assumption and instruction, multiple and intersectional discrimination, victimisation, segregation, hate crimes and hate speech). A similar scoring method is used for the number of areas of life regulated by the law (where a score of 1 is given if employment, education and training, health, social services, justice, online and media, provision of and access to products, and housing are covered by anti-discrimination laws). Scores of 1 are then given if positive actions are permitted. These scores are then summed and divided by 3 to create an overall score between 0 and 1, where 0 means there is no legal protection against discrimination and a score of 1 means all forms of discrimination are prohibited in all areas of life considered, as well as whether positive actions are permitted. For the disability ground, the law index also includes whether the law requires reasonable accommodations (0 for no, 1 for yes). This score is averaged with the other components for an overall score between 0 and 1.
The policy index assesses government initiatives aimed at addressing the needs of groups at risk of discrimination. It examines various types of initiatives that sit within the policy areas of employment, education and training, health, justice and safety, disability accessibility and awareness campaigns. The types of initiatives that are considered within each policy area are outlined in the light-blue boxes under the policy index part of Annex Figure 3.A.1. Each broad policy area receives a score between 0 and 1, where 1 is given if a country has examples of all the types of initiatives that are covered. For example, a country will score 1 for health policy if they have community-based programmes, targeted health campaigns and tailored care in mainstream settings, and 2/3 if they have two of these three initiatives. Scores for the broad policy areas are then averaged to create an overall policy score.
The mainstreaming index evaluates efforts to mainstream non-discrimination and equality throughout government processes. It assigns scores based on whether countries have mechanisms like coordination processes, national equality strategies, inclusive stakeholder engagement, and training for public officials (depicted in the light blue boxes of Annex Figure 3.A.1). For some mechanisms, countries receive a point if they have them and 0 otherwise (such as whether coordination processes are in place). In the case of equality strategies, countries receive a point if their strategy considers intersectionality and structural discrimination, which is best practice (European Network Against Racism, 2019[74]). Countries receive a score of 1/3 for having an equality strategy without either of these features, 2/3 if they have one of these elements and 0 if they do not have an equality strategy. Similarly, for data collections and staff training, scores account for scope. Countries receive a score of 1 if they collect data in all areas of interest (e.g. employment, education and training, health and justice) and train staff in all key policy areas. Fractional scores are given for the share of areas covered by equality data collections and staff training, and 0 is given if no data are collected on at-risk groups and staff training is not offered. The scores for all mainstreaming mechanisms are then averaged to obtain an overall mainstreaming score between 0 and 1.
Finally, the law, policy and mainstreaming indices are averaged to create an overall index for each protected ground. The three sub-indices are given equal weight in the overall index, consistent with the approach used in previous OECD efforts to build anti-discrimination and inclusion indices (e.g. the OECD (2020[11]) LGBTI-Inclusivity Index). Scores closer to 1 indicate that countries have strong legal, policy and mainstreaming efforts for the ground specified, while a score closer to 0 suggests weaker protections.
Source: Based on the OECD Anti-Discrimination Questionnaire.
← 1. A note on terminology: in this chapter, inclusive language is used as much as possible when referring to LGBTI people, persons with disabilities and people from racialised communities. The term “racialised communities” is based on the European Commission Against Racism and Intolerance’s (2021[75]) definition, in which racialisation is “the process of ascribing characteristics and attributes that are presented as innate to a group of concern to it and of constructing false social hierarchies in racial terms and associated exclusion and hostility. Regardless of where one is from and of personal circumstances, once identified or perceived as a member of a group, one is deemed as embodying characteristics based on, for instance, skin colour, ethnic or national origin inherent to all members of that group”. The terms “at-risk groups” and “minority groups” are used interchangeably. However, in some cases, the chapter deviates from using LGBTI in order to accurately present studies on population subgroups (e.g. LGBT for lesbian, gay, bisexual or transgender individuals or LGB for lesbian, gay and bisexual people).
← 2. The other surveys analysed in this chapter – 2023 Discrimination in the EU Eurobarometer and AXA Mind Health – do not define discrimination in the framing of the questions on people’s experiences of discrimination, which makes it difficult to test whether respondents interpreted the questions in the same way. However, the results presented in Chapter 2 indicate that people had similar interpretations, as the self-reported discrimination rates across these surveys and the Opportunities Module of the 2022 OECD Risks that Matter Survey were closely aligned.
← 3. The 2023 Discrimination in the EU Eurobarometer indicates that 13% of respondents who experienced discrimination in the past year stated that it was due to their socio-economic status.
← 4. These results are consistent with qualitative research from the United Kingdom, which indicates that loneliness felt by people from Black, Asian and Ethnic Minority (BAME) communities stems from being treated with less respect or courtesy because of their ethnicity (in 46% cases) and religion (in 49%) (British Red Cross, 2019[36]). Loneliness rates were higher among BAME-identifying people who stated that they did not feel like they belonged, felt unsafe or had experienced discrimination or racism. Almost half of people who had experienced discrimination at work or in their local neighbourhood reported being always or often lonely, compared with just over a quarter of people who had not experienced discrimination (British Red Cross, 2019[36]).
← 5. Estimates are calculated by considering a wage gap using a 3% LGBT incidence rate and a 15% loss of productivity, and by attributing a cost to the excess prevalence rate due to LGBT discrimination.
← 6. The estimated costs for each group at risk of discrimination are uplifted to 2022 prices using Eurostat’s Harmonised Consumer Index of Prices.
← 7. The OECD Anti-Discrimination Questionnaire was completed by 21 OECD EU countries, comprising Austria, Belgium, Czechia, Denmark, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Latvia, Lithuania, Luxembourg, the Netherlands, Poland, Portugal, Slovak Republic, Slovenia, Spain and Sweden.