Race, ethnicity, and discrimination at work: a new analysis of legal protections and gaps in all 193 UN countries

Discrimination in hiring

A wide range of studies have demonstrated racial and ethnic discrimination in hiring, including studies in which researchers submit fictitious CVs and applications that reflect similar credentials and experience, but that vary with respect to photos, names, and/or experiences suggestive of different racial or ethnic identities. These “correspondence studies,” which improved on prior methods of testing for racial discrimination by making candidates substantively identical except for markers of race/ethnicity (Bertrand and Duflo, 2017), find that presumed race/ethnicity influences the likelihood that a particular candidate receives an invitation to interview, with those representing historically marginalized racial or ethnic groups consistently receiving fewer callbacks (Baert, 2018).

Other research approaches include direct interviews with hiring managers and simulations in which study participants rate the strength of hypothetical job candidates based on their photos and descriptions of their experience and characteristics where, again, the principal aspect varied is race/ethnicity, either on its own or together with intersectional characteristics like migration status or gender.

These research approaches also document the persistence of discrimination in hiring across jobs and geographies. For example, research in Nigeria found that managers of both public and private organizations were more likely to hire applicants from their own ethnic group (Adisa et al., 2017). A study spanning five European countries—Germany, the Netherlands, Norway, Spain, and the United Kingdom—demonstrated discrimination in the hiring of Black and Middle Eastern men (Di Stasio and Larsen, 2020).

Discrimination based on common proxies for race or ethnicity can likewise shape job prospects. In Canada, for example, migrants from sub-Saharan Africa report that their accents can be a barrier to becoming employed and having career mobility (Creese, 2010), while in the US, numerous court cases have illustrated how Black women commonly face barriers to employment because their natural hairstyles are found to violate “neutral” grooming codes (Greene, 2017).

Discrimination is also often intersectional. In Germany, a 2020 study found that women with Turkish names were less likely than those with German names to receive interview invitations, and this gap widened further for women wearing headscarves (Weichselbaumer, 2020). Similarly, in a Mexico study, both marital status and skin color affected women's chance of receiving an interview (Arceo-Gomez and Campos-Vasquez, 2014). In Belgium, women from minority ethnic groups were less likely to be considered for a “high-cognitive demanding job” than either native women or minority ethnic men (Derous and Pepermans, 2019).

Discrimination in promotions

Studies have also documented racial and ethnic discrimination in promotions across professions, from police forces to law firms to universities (Tomlinson, 2019; Zempi, 2020). From Finland to South Africa to the United Kingdom and the United States, workers from marginalized racial and ethnic groups report discrimination in promotion, consistent with the research evidence based on multilevel multivariate studies of discrimination, as well as based on implicit bias testing of supervisors (Hatch et al., 2016; Mayiya et al., 2019; Stalker, 1994; Yu, 2020; Zempi, 2020). In Canada, research has documented that visible minorities have less upward mobility even after controlling for education, work experience, time with the employer, and other factors (Javdani, 2020), including both supply- and demand-side factors (Javdani and McGee, 2018; Yap, 2010; Yap and Konrad, 2009).

Aside from direct discrimination in promotions, employer practices that evaluate employee conduct differently or otherwise deny opportunities for professional advancement based on race or ethnicity can affect opportunities within the workplace. For example, a study that experimentally changed the race/ethnicity of an employee in a photo while asking study participants to evaluate their performance demonstrated that simple acts such as being late for work led to a significantly greater negative impact on the appraisal of hypothetical employees when the photo showed a Black or Latinx employee than when the photo showed a white employee (Luksyte et al., 2013). Visible minorities are also less likely to receive training opportunities that can influence upward mobility in the labor force (Dostie and Javdani, 2020).

Discrimination in terminations

Both direct discrimination by employers and structural discrimination that cuts across economies can make workers from marginalized racial and ethnic groups more vulnerable to terminations. For example, studies have found that during economic downturns, immigrants and workers from historically marginalized racial and ethnic groups face heightened risks of labor market discrimination and job loss (Couch and Fairlie, 2010; Lessem and Nakajima, 2019). Moreover, the consequences of past discrimination and exclusion from economic opportunities mean that workers from underrepresented groups are less likely to have seniority within a given organization or company. As a result, in addition to direct racial/ethnic discrimination that may lead to higher rates of termination, “last hired, first fired” policies can result in indirect discrimination against workers from historically excluded groups.

Impacts of discrimination in hiring, job positions, and promotions on pay inequality

Discrimination in hiring can impact initial salaries and level and type of starting position. When individuals are hired into jobs below their skill level because of bias based on race and ethnicity, they earn less than they would have earned had there been no discrimination (Coleman, 2003). Likewise, when discrimination results in the overrepresentation of workers from historically marginalized racial/ethnic groups in limited employment capacities, including temporary or seasonal jobs, gaps in both pay and benefits further widen. Survey research across 30 European countries showed that even after controlling for education, experience, occupation, and other categories, racial and ethnic minorities were more likely to end up in jobs where their skills were underutilized, leading to lower wages than if they were in a position more matched to their skills and offering reduced pathways for advancement (Rafferty, 2020). In Chile, qualitative research has found that Peruvian migrants simultaneously experience limited employment trajectory due to their external migrant status alongside racialization by local Chileans who perceive them to be more fit for low-status and low-income positions due to assumptions about their physical and cultural traits (Mora and Undurraga, 2013).

Direct pay discrimination

Even for the same job position, the “unexplained” wage differential after taking experience into account gives an indication of the amount of the wage differential that could be due to discrimination and bias. One-half to two-thirds of wage differences across racial and ethnic groups in some studies have been estimated to be due to bias (Drydakis, 2012; Piazzalunga, 2015). While the data clearly demonstrates the existence of bias and discrimination in pay against specific groups in a range of countries, there has not been a comprehensive look across countries and racial/ethnic groups to document in detail when and where the wage gaps are greatest and lowest, before and after taking into account the impact of bias throughout the work lifecourse.

The documented and potential impacts of national laws addressing discrimination

Individual countries that have passed antidiscrimination laws have seen improvements including greater equality in hiring and lowering of wage disparities (Leck et al., 1995). While antidiscrimination laws alone do not eliminate discrimination in hiring, pay, promotions, or terminations, studies both across countries and across populations have demonstrated that antidiscrimination laws can make a difference. In Canada, for example, studies of the Employment Equity Act found that the share of visible minorities who were employed in the private sector increased to much closer to the percentage of the population following the law's adoption (Agocs, 2002; Leck and Saunders, 1992). In the United States, studies have found that antidiscrimination laws contributed to wage and income increases for Black workers (Collins, 2003; Donohue and Heckman, 1991) and a narrowing of the racial/ethnic pay gap (Chay, 1998).

These findings on laws' impacts on employment outcomes by race parallel those observed for other groups of marginalized workers. For example, one study of 141 countries found that laws prohibiting gender discrimination in employment increased women's labor force participation in formal jobs (del Mar Alonso-Almeida, 2014), while in the UK, legislation guaranteeing equal pay and non-discrimination in employment on the basis of sex resulted in a 19.4% increase in women's earnings and a 17% increase in women's employment rates relative to men's (Zabalza and Zafiris, 1985). Moreover, explicitly prohibiting all forms of workplace discrimination matters to norms. In addition to their practical or applied value, laws prohibiting discrimination have important expressive value that can shape workplace expectations as well as societal views of equality more broadly, with the potential to affect rates of both explicit and implicit bias (Sunstein, 1996). At the same time, the past several decades of antidiscrimination law have revealed important gaps to address. First, as many of the studies cited in the previous section illustrated, racial and ethnic discrimination commonly co-occurs with discrimination based on migration status, foreign national origin, social class, and other characteristics, highlighting the cumulative and often intersectional impacts of key facets of identity on work-related experiences around the world. Clearly banning all common grounds of discrimination, including those used as proxies for race or ethnicity or that commonly intersect with race or ethnicity, is a critical first step.

Second, prohibitions of indirect discrimination can offer important protection against racial/ethnic discrimination, including in instances where discrimination based on an unprotected ground has disparate impacts on the basis of race or ethnicity. This is true both for common grounds of discrimination that would ideally be explicitly covered by domestic labor laws (as they are by international treaties, e.g. national origin) (Demetriou, 2016), as well as proxies for racial/ethnic discrimination that are not generally addressed on their own (e.g. accents and hairstyles) (Justesen, 2016). In contrast, when discrimination laws take an overly formal approach to discrimination that only covers acts that were direct or intentional, they fail to account for the extensive evidence demonstrating that policies and practices that are racially neutral on their face may have disproportionate consequences for workers from historically marginalized groups.

Third and finally, while protections against employment discrimination are essential, more attention must be paid to implementation. While a range of actions are needed, evidence shows that having legal protections in place against retaliation may increase reporting rates by reassuring workers that their careers will be protected if they report discrimination (Bergman et al., 2002; Gorod, 2007; Keenan, 1990; Pillay et al., 2018).

This is the first study to examine legislation in all 193 UN countries to map the extent to which each country in the world has protections against racial and ethnic discrimination in hiring, promotions, training, demotions, and terminations, as well as whether they proactively support implementation through clear legislative prohibitions of retaliation for reporting. Further, we examine to what extent countries not only address direct discrimination based on race/ethnicity, but also indirect racial/ethnic discrimination and/or direct discrimination based on grounds that can serve as proxies depending on the historical and societal context for racial discrimination, including religion, migration status, and socioeconomic status. Further, we highlight examples where countries explicitly address intersectionality. Finally, we examine whether there were gains over the past five years in the number of countries that are prohibiting each type of discrimination.

Data source

We constructed a database of prohibitions against discrimination in private sector labor in all 193 UN member states as of January 2021. Labor, anti-discrimination, and other relevant pieces of legislation were identified through the International Labor Organization's NATLEX database, supplemented with legislation identified through country websites. A coding framework was developed to systematically capture key policy features. This coding framework was reviewed by researchers, lawyers, and other leaders working on employment discrimination and tested on a subset of countries before database coding commenced.

For each country and protected characteristic studied, two researchers independently read legislation in its original language or a translation and used the coding framework to assess whether legislation specifically prohibited discrimination in each aspect of work or broadly, whether there were any exceptions to prohibitions of discrimination based on employer characteristics, and whether there were specific provisions in place to support effective implementation. In countries where anti-discrimination protections are legislated subnationally, the lowest level of protection across states or provinces was captured. Answers were then reconciled to minimize human error. When the two researchers could not arrive at an agreement based on the codebook framework, the full coding team met to discuss, and the coding framework was updated to reflect the decision. When updates were made, countries that had already been coded were checked for consistency with the update.

Once coding was complete, systematic quality checks were conducted of variables that proved challenging for researchers during the coding process. Randomized quality checks were conducted of variables that were more straightforward, checking first twenty countries to ensure no errors were identified and a larger subset of countries if there were errors. Finally, outlier verifications globally and by region or country income level were conducted for all variables. In order to assess whether legislative provisions have strengthened over time, similar methods were used to construct measures of laws in place as of August 2016.

Variables

Analysis

All analyses were conducted using Stata MP 14.2. Differences were assessed by region using the Pearson's chi-square statistics. Region was categorized according to the World Bank's country and lending groups as of 2020 [1].

Overall

Globally, 153 countries prohibited at least some form of racial and/or ethnic discrimination at work in 2021, a modest increase from 148 countries in 2016 (Figure 1). Three of the countries introducing these new prohibitions were in Sub-Saharan Africa (Mali, South Sudan, and Zambia), one in Europe (Iceland), and one in the South Pacific (Tuvalu). An additional five countries expanded existing prohibitions of racial and/or ethnic discrimination either to broadly prohibit discrimination at work in addition to specific prohibitions in certain areas (Barbados and Honduras) or to comprehensively cover discrimination at work in all areas, as well as indirect racial and/or ethnic discrimination (Andorra, Burundi, and Sao Tome and Principe).

Gaps in prohibitions are found in every region of the world. Countries in the Americas were the most likely to prohibit at least some form of racial discrimination at work, followed closely by Europe and Central Asia and sub-Saharan Africa. In each of these regions, only ten percent or fewer of countries lacked at least some form of prohibition. In contrast, a majority of countries lack prohibitions of racial discrimination in East Asia and Pacific and South Asia (Figure 2). Differences were statistically significant between these two regions and the three regions with the highest levels of prohibitions (p < 0.01).

Hiring

In 2016, 107 countries had a law that explicitly prohibited race-based discrimination in hiring. That number increased to 115 countries in 2021 (see Figure 2). An additional 27 countries in 2016 and 29 countries in 2021 had either a broad prohibition of race discrimination or a general prohibition of discrimination in hiring. Prohibitions of racial/ethnic discrimination in hiring were most common in Europe and Central Asia (91%) followed by sub-Saharan Africa (62%). In all other regions, fewer than half of countries prohibited racial discrimination in hiring (Figure 3).

Training and promotions/demotions

Eighty countries in 2016 and 88 countries in 2021 prohibited discrimination based on race in training. Eighty-three countries in 2016 prohibited discrimination in promotions and demotions. In 2021, this number increased to 90 countries.

While less than three-quarters of countries prohibited racial discrimination in training in Europe and Central Asia (74%), these prohibitions were still more common than in every other region. Similar trends were found for promotions and/or demotions.

Pay

The number of countries guaranteeing equal pay for work of equal value, increased from 34 in 2016 to 41 in 2021. Overall, including guarantees both of equal pay for work of equal value and equal pay for equal work, 91 countries in 2016 and 96 countries in 2021 guaranteed equal pay across racial and ethnic groups. More countries in Europe and Central Asia (77%) prohibited racial discrimination in pay than those in sub-Saharan Africa (56%), the Middle East and North Africa (27%), and South Asia (27%).

Terminations

In 2016, 106 countries prohibited racial discrimination in terminating employment. That number increased to 112 countries in 2021. Prohibitions of racial/ethnic discrimination in terminations were most common in Europe and Central Asia (74%), Sub-Saharan Africa (71%), and the Americas (60%). In contrast, less than a third of countries prohibited racial discrimination in terminations in East Asia and Pacific (30%) and South Asia (25%).

Harassment

Only 66 countries explicitly prohibited workplace harassment based on race/ethnicity in 2016. By 2021, that number had increased to 72. Only a minority of countries prohibited racial and/or ethnic harassment in all regions except Europe and Central Asia.

Indirect discrimination

Sixty-three countries prohibited indirect discrimination based on race and/or ethnicity in 2016, increasing to 71 countries in 2021. Only a third of countries in sub-Saharan Africa, a fifth of those in East Asia and Pacific and the Americas, and an eighth of those in South Asia explicitly addressed indirect racial/ethnic discrimination. No countries in the Middle East and North Africa did so.

Intertwined, multiple and intersectional discrimination

Prohibitions of discrimination based on both race and/or ethnicity and religion were widespread: 151 countries prohibited at least some aspect of workplace discrimination based on both characteristics in 2021. However, only 117 countries prohibited at least some aspect of workplace discrimination based on both race and/or ethnicity and social class. Even fewer prohibited discrimination based on both race and/or ethnicity and foreign national origin (90 countries) or migration status (81 countries).

Countries in sub-Saharan Africa were most likely to prohibit discrimination based on race and social class, as well as discrimination based on race and foreign national origin. While prohibitions of racial discrimination and religion, social class, or foreign national origin were comparatively high in the Americas, prohibitions of discrimination based on migration status were markedly lower. While nearly two-thirds of countries in Europe and Central Asia addressed migration status alongside race, only half prohibited discrimination based on foreign national origin (Figure 4).

Finally, a minority of countries explicitly addressed the concepts of intersectionality or multiple discrimination in their discrimination legislation. Kenya's National Gender and Equality Commission Act recognizes intersectionality in defining marginalized groups to be people “disadvantaged by discrimination on one or more of the grounds in Article 27(4) of the Constitution” which includes “race, sex, pregnancy, marital status, health status, ethnic or social origin, colour, age, disability, religion, conscience, belief, culture, dress, language or birth” (2012). Australia's Racial Discrimination Act prohibits “acts done for 2 or more reasons” where “one of the reasons is the race, colour, descent or national or ethnic origin of a person” (1975). In Macedonia, the Law on Prevention of and Protection Against Discrimination defines multiple discrimination to be a severe form of discrimination (2010).

Employer responsibilities

Only a smaller minority of countries (38) took the additional step of requiring employers to take one or more specific measures to prevent racial discrimination. An additional 8 countries had general language requiring employers to take preventative steps, without specifying what those steps would look like.

Protections in the event of discrimination

In the event that discrimination occurred and employees filed a report or initiated litigation, a modest majority of countries took the important step of prohibiting retaliation against the employee who filed the complaint. Seventy-eight countries prohibited employers from retaliating in any way, an additional 7 prohibited harassment or any disciplinary action, and 26 only prohibited dismissing the employee (Table 1). A similar number of countries (76) protected employees who participated in investigations from being retaliated against.

Are any employers exempt?

When countries had laws in place prohibiting discrimination, they overwhelmingly applied to all employers. In rare cases small businesses were exempt, including in 5 countries in the case of hiring, 4 countries in the case of training and terminations, and 3 countries in the case of pay and promotions and demotions (Table 2). Charities and nonprofits had similarly uncommon exemptions. The group that was most frequently exempted from these prohibitions of racial discrimination were religious organizations. Fourteen countries exempted religious organizations from bans on discriminations based on race in hiring, training, and terminations, 13 exempted religious organizations in terms of racial discrimination in promotions and demotions, and 9 in the case of pay.

The critical need to accelerate progress

The significant overall gaps in protections, alongside the findings that the expansion over the past five years of laws prohibiting racial discrimination at work has been slow, underscores the need to accelerate the pace of change on legal reforms—as a matter of human rights, an important determinant of individual and household incomes, and a prerequisite for countries to reach their full potential. Ensuring equal opportunities in employment on the basis of race and ethnicity has vast implications for individuals, families, and their broader communities. A significant body of literature has documented how access to employment, job quality, and adequate income shape mental and physical health, overall life satisfaction, and the ability to meet material needs (Calvo et al., 2015; Murphy and Athansou, 1999). When work opportunities are unevenly distributed by race due to both individual and structural discrimination, these disparities drive broader inequalities.

Moreover, all countries have agreed to do so. The Sustainable Development Goals, adopted by all UN member states in 2015, commit governments to “empower and promote the social, economic and political inclusion of all, irrespective of age, sex, disability, race, ethnicity, origin, religion or economic or other status” and “[e]nsure equal opportunity and reduce inequalities of outcome, including by eliminating discriminatory laws, policies and practices and promoting appropriate legislation, policies and action in this regard” (United Nations Department of Economic and Social Affairs, 2015). By adopting the SDGs, countries worldwide agreed to realize these commitments by 2030. To meet that timeline, accelerating the pace of change on fundamental anti-discrimination protections is essential.

This builds on a long history of international agreements guaranteeing equal rights regardless of race or ethnicity, including in the field of employment. Many of these have been in place for decades. Foremost among them is the International Convention on the Elimination of All Forms of Racial Discrimination (United Nations Office of the High Commissioner for Human Rights, 1965), which has been ratified by 182 countries (United Nations Office of the High Commissioner for Human Rights, 2021c), declares that States Parties have a duty to “prohibit and to eliminate racial discrimination in all its forms and to guarantee the right of everyone, without distinction as to race, colour, or national or ethnic origin, to equality before the law, notably in the enjoyment of (…) The rights to work, to free choice of employment, to just and favourable conditions of work, to protection against unemployment, to equal pay for equal work, to just and favourable remuneration” (United Nations Office of the High Commissioner for Human Rights, 1965).

In fact, nearly every major global human rights agreement commits countries to treating all people equally regardless of race or ethnicity. These include, among others, the Universal Declaration of Human Rights, adopted as the first global agreement of the United Nations and considered binding on all countries (United Nations General Assembly, 1948), the International Covenant on Economic Social and Cultural Rights, a binding treaty ratified by 171 countries (United Nations General Assembly, 1966; United Nations Office of the High Commissioner for Human Rights, 2021b), and the International Covenant on Civil and Political Rights, ratified by 173 countries (United Nations Office of the High Commissioner for Human Rights, 1966, 2021a).

Finally, beyond its importance to individuals, families, and communities and deep intrinsic value as a matter of human rights, ending racial discrimination in the labor market has significant implications for economies and companies. For example, in the US alone, estimates from the Federal Reserve Bank of San Francisco find that closing the racial gaps in employment-to-population ratios between 1990 and 2019 would have boosted 2019 GDP by over $150 billion (Buckman et al., 2021), while other research has forecast that closing the racial earnings gap by 2050 would boost GDP by 22% (Turner, 2018). Likewise, a significant body of evidence demonstrates that greater racial and ethnic diversity within companies, including on boards, improves their financial performance and degree of innovation (Erhardt et al., 2003; Herring, 2009; Cheong and Sinnakkannu, 2014; Thomas et al., 2016; Hunt et al., 2018). For example, a study of 492 firms found a strong relationship between ethnic and linguistic diversity and total revenue, dividends, sales and productivity (Churchill, 2019). Particularly as more companies work and hire trans-nationally, the extent to which laws in all countries prohibit racial and ethnic discrimination at work matters to overall performance.

Research limitations and the need for a broader research agenda on policies and outcomes

While this study provides an important first look at prohibitions against racial and ethnic discrimination at work in all the world's countries, it has important limitations. This study did not quantify laws related to intersectional discrimination, including, among others, gender, age, disability, sexual orientation, and gender identity. Yet as past scholarship and case law have shown, addressing each individual basis for discrimination still may not be enough to reach the unique forms of discrimination that arise when multiple grounds of discrimination intersect, particularly if workers are required to prove each of their discrimination claims discretely and sequentially. Explicit protections against intersectional discrimination, and judiciaries willing and trained to apply them, may be needed (Crenshaw, 1989; Fredman, 2016).

Future research should also examine prohibitions of discrimination in working conditions, given the evidence of inequalities. For example, research has shown that non-Hispanic Black workers and foreign-born Hispanic workers are disproportionately hired into jobs with the higher injury risk and increased prevalence of work-related disability (Seabury et al., 2017). Another study on COVID-19 job exposures found that Latinx and Black frontline workers were overrepresented in lower status occupations associated with higher risk and less adequate COVID-19 protections, contributing to the higher prevalence of infection in these populations (Goldman et al., 2021).

Further, we need to measure laws that reduce bias in informal as well as formal mechanisms that play a large role in the recruitment, hiring, and promotion processes, as well as in determining working conditions. Evidence has shown social networks and informal relationships can not only impact recruitment, but also can contribute to inequities in salary negotiations and mentorship at the hiring stage (Seidel et al., 2000; Spafford et al., 2006).

These expansions on the law and policy data presented here should be part of a broader research agenda on racial equity in the global labor market that examines not only which laws and policies are in place but what impacts they are having. As with other policy areas, developing longitudinal quantitative, globally comparative measures of anti-discrimination laws helps make it possible for researchers to rigorously analyze the relationship between policy change and outcomes, producing actionable evidence about “what works” across countries (Raub et al., 2022). However, even with the new policy data we have developed, improvements in outcomes data will be essential to measure the impact globally of advances and legal gaps. Globally comparative data on experiences of racial discrimination across countries which is essential for measuring the impact of legal change globally has been limited to date for several reasons including the wide range across countries of who suffers racial and ethnic discrimination and the variability of country willingness to collect data.

Addressing discrimination: a global responsibility

While the workplace is only one location where racial and ethnic discrimination occurs, it is a crucial one. Ensuring equal opportunity to be hired and equal treatment in pay, working conditions, and promotions together influences whether individuals can lead full work lives, contribute to household income, and not only meet basic needs but also invest in the future of their families and communities. Moreover, global agreements have committed countries around the world to combatting discrimination based on race and ethnicity, including in the specific context of employment.

For these international instruments to have full impact, however, specific country-level legislation that operationalizes their commitments and prohibits discrimination in the workplace is essential. To accelerate progress toward ending racial and ethnic discrimination in employment worldwide—a basic human right—we need to monitor the steps countries are taking to address and eliminate discrimination in all aspects of work, including hiring, promotion, pay, and terminations. While legal guarantees are not enough—and norm change, leadership, and social movements are likewise critical to successfully eliminating discrimination both at work and more broadly—clear stipulations that companies are not allowed to discriminate are essential, as are strong and regularly updated accountability mechanisms.