This chapter reviews the main theories developed to explain discrimination. Mirroring the historical development of the field, while reflecting a theoretically systematic approach (Pager and Shepherd 2008; Reskin 2003), the chapter adopts an approach by analytical scales to present and discuss theories of discrimination. The first section presents theories seeking the cause of prejudice and discrimination at the individual level, the second section focuses on organizational mechanisms and the third on structural determinants.

3.1 Individual-Level Theories

Defined as a behavior or a decision based on ascriptive characteristics such as race or ethnic background, discrimination differs from stereotypes and prejudices, which are mental representations summarizing the evaluation of groups. Stereotypes represent the cognitive component of such mental representations or attitudes, while prejudices describe the affective component at the roots of a biased behavior disadvantaging individuals based on their group membership or minority position. In the words of Gordon Allport, a stereotype is “an antipathy based on faulty and inflexible generalization. It may be felt or expressed. It may be directed toward a group as a whole or toward an individual because he is a member of that group” (1954, 9). Yet, attitudes are at the core of individual-level explanations of why discrimination occurs. As such, they are prominently discussed in this first section.

3.1.1 Individual Psychological Conflicts

Early theories located the motives for discrimination in the character and personality of individuals (Fiske 1998). In this perspective, internal motivations of actors are seen as rooted in individual psychological conflicts and in intrapsychic factors, such as negative attitudes against minority groups. Adorno’s theory of the authoritarian personality (Adorno et al. 1950) is iconic for highlighting intrapsychic factors as causes of blatant discrimination. Echoing Freud’s psychoanalysis, this theory argues that individuals inclined to conservatism, nationalism, and fascism tend to develop a rigid personality, think in rigid categories, express conventional beliefs, and often identify with and submit themselves to authority figures. According to Adorno, individuals with authoritarian personalities develop aversion toward differences to their own values and norms and thus express an overt negative attitude toward minority groups.

Though very prominent in the 1950s, the authoritarian personality theory, in its original form, is today considered outdated, notably because it fails to account for observed changes in prejudice and discrimination over time. However, in the field of political psychology, there has recently been a renewed interest in this theory (Funke et al. 2016). The association between authoritarianism and prejudice indeed seems to be driven by collective rather than an individual threat (Pettigrew 2016).

In the 1960s, conceptualization of prejudice gradually changed. While it used to be understood as a psychopathological expression among traditionally minded, conservative, and educationally disadvantaged individuals, it increasingly came to be seen as rooted in socio-psychological processes of social cognition, group dynamics and socialization among ordinary people (Dovidio et al. 2010; Dovidio 2001). With the rise of the civil rights movement and the ensuing promotion of non-discrimination (Civil Rights Act of 1964 in the US and the Race Relation Act of 1965 in the UK), overt expressions of prejudice declined (Schuman et al. 1997). However, it was supplanted by subtle forms of discrimination, consistently observed in North America and in a number of Western European countries (Pettigrew and Meertens 1995). Such subtle discrimination is characterized by ambivalence: majority group members may publicly profess equality while still holding negative attitudes toward minority members in the private sphere, and biases against out-groups might even be implicit or unconscious. They express themselves in non-verbal behavior, less friendly attitudes in interaction with minority groups and aversion toward them (Dovidio et al. 2002).

A range of theories, mainly deriving from the US context, emphasized this transition from overt to more covert or subtle forms of discrimination, such as symbolic racism (Sears and Henry 2003) and modern racism (McConahay 1986). Both of these theories take as their point of departure the conflicting and often ambivalent attitudes of majority group members: humanitarian sympathy for underprivileged persons often goes hand in hand with the blaming of the victims for failing to comply with individualistic values. In this perspective, minority members are resented as they are deemed to ostensibly disregard traditional conservative values (e.g., a Protestant work ethic) and to make unjustified and excessive claims. Conservatism manifests itself with support for the existing power relations in society and with opposition to policy measures in favor of minority groups.

Aversive racism theory (Gaertner and Dovidio 1986) also deals with subtle, ambivalent attitudes, but focuses on the ambiguities among liberal-minded majority members. While professing equality, those majority individuals still hold conflicting, non-conscious negative feelings about minorities; the resulting discomfort, anxiety, and fears lead to an aversion of contact. Consistent with their non-racist self-image, liberal-minded majority individuals refrain from acting in overtly discriminatory ways; yet, coherent with their unconscious negative attitudes resulting from socialization, they are likely to avoid situations where they come into contact with members of minority groups and tend to refrain from supporting equalizing policies.

Contemporary, subtle forms of discrimination rest on the dissociation between inclusive egalitarian attitudes and unconscious pervasive bias, between controlled responses and automatic responses that can be attributed to immediate associations with an evaluative content. Implicit biases may operate unconsciously to influence behavior. This dissociation model stimulated important methodological developments (Greenwald et al. 1998), suggesting that self-report methods are appropriate for the measure of explicit attitudes but unsuitable for implicit attitudes. Indeed, this research has demonstrated that self-reports and implicit measures of stereotyping and prejudice are largely uncorrelated (Dovidio et al. 2015, 5).

The subtle character of contemporary bias and the impact of implicit attitudes are further at the roots of theories of “color-blind racism” (Bonilla-Silva 2003). To address the effects of implicit bias, well-meaning majority people may emphasize common group identity in a color-blind approach to diversity: they treat individuals as equally as possible, without considering their race, culture, or ethnicity, in order to foster positive intergroup relations. However, common group identity is related to color-blind assimilation ideologies, so that the minority group is expected to conform to dominant norms and values. Color-blind policies tend to preserve white privilege and to maintain minority disadvantages. Stressing color-blindness proves to be a strategical tool: it reinforces hierarchical relations between groups, benefiting high-status majority group members. The other downside of this frame is that it limits awareness of social inequalities, thus it might hamper effective action to address those issues through social change.

3.1.2 Individual-Level Factors in the Labor Market: The Rationale of Gatekeepers

Much research on discrimination aims at understanding the role of differential treatment in the marketplace, such as labor markets, housing markets or the consumer markets (see Chap. 5). While psychologists have approached such market discrimination with the study of stereotypes and attitudes, economists have developed specific theoretical frames to account for discrimination, distinguishing between taste-based and statistical discrimination. In his seminal book, The Economics of Discrimination (1957), Becker, for example, discusses the economic effects of racial discrimination in the US labor market. In this book, Becker defines overt racism as individuals’ aversion for interracial contact and qualifies it as a “taste” for discrimination. According to Becker, racial discrimination is the result of employers’ willingness to pay for not being associated with African Americans – either by rejecting the most productive candidates or by offering a reduced income. In this theoretical model, discrimination is explained with reference to direct racial animus among employers because the behavior lacks “objectivity.” Rational behavior is deemed to be based on considerations about productivity alone, and discrimination is thus a result of employers acting based on subjective preferences. As such, an underlying assumption in Becker’s theory is that discriminatory employers over time will be crowded out of the labor market because their behavior lowers productivity.

In contrast to the assumption that discrimination and productivity are mutually exclusive, economic models of statistical discrimination, originating from the work of Phelps (1972) and Arrow (1973), rest on the idea that discrimination is a way of managing the imperfect information that characterizes hiring decisions and wage setting in the labor market. According to Phelps, “the employer who seeks to maximize expected profit will discriminate against blacks or women if he believes them to be less qualified, reliable, long-term, etc. on the average than whites and men, respectively, and if the cost of information about the individual applicants is excessive” (Phelps 1972, 659). In the absence of full information, race, ethnicity, and sex will be used as proxies for productivity. According to this theory, risk-aversive employers will hire the candidate who is ascribed membership to the group that has the highest average productivity – presumably whites and men.

The main difference between taste-based and statistical discrimination is the notion of rationality (Midtbøen 2014). Excluding the most productive job applicant on the grounds of race or sex is economically inefficient, while hiring decisions based on estimates of group productivity are assumed to be rational (although still discriminatory) responses to the uncertainty and lack of full information characterizing hiring decisions in the labor market. The employer may reject a suitable candidate because of statistical discrimination, but this cost is traded off against the cost of (trying) to find out the real productivity of all candidates. Both uncertainty and lack of information are inevitable parts of recruitment processes, and a characteristic of organizational behavior as such (Stinchcombe 1990). Nevertheless, an unclear aspect of statistical discrimination models is the question of accuracy in employers’ beliefs about average group productivity, which relies heavily on stereotypes. Both Phelps (1972) and Arrow (1973) are somewhat vague on this point, indicating – perhaps – that their models allow for employers’ beliefs about blacks and women to be inaccurate depictions of reality and still be “rational” in some sense. Statistical discrimination might thus involve some sort of racist beliefs, even though employers do not consider that they mobilize stereotypes against ethno-racial minorities or women.

To clarify this point, other economists define statistical discrimination as a situation where employers act on the basis of “true stereotypes” (Schwab 1986, 228), arguing strongly that average differences in productivity between whites and blacks, or between men and women, actually exist on average and that this difference is the basis of discrimination (Aigner and Cain 1977). Moreover, an entire branch of the economics literature is concerned with so-called employer learning (e.g., Altonji and Pierret 2001; Farber and Gibbons 1996). These scholars acknowledge that statistical discrimination may be based on outdated beliefs about group productivity, but argue that employers who have positive experiences with stigmatized minority groups will update over time their beliefs to be in accordance with empirical realities (Farmer and Terrell 1996). By effect of a similar learning process, economists would assume that in the long-term employers would better master how to identify the productive candidates, thus reducing statistical discrimination (Midtbøen 2014).

Many sociologists have criticized economic models of statistical discrimination, questioning the idea of accuracy in beliefs about group productivity (e.g., Bielby and Baron 1986; Tomaskovic-Devey and Skaggs 1999), along with the assumption that employers update their views of racial minorities when new and positive information is provided (Pager and Karafin 2009). The idea that employers are guided by “true stereotypes” stands, for example, in striking contrast to the definition of prejudice as “an exaggerated belief associated with a category” (Allport 1954, 191; Fiske 1998). Indeed, important qualitative work both in the US context (e.g., Kirschenman and Neckerman 1991; Moss and Tilly 2001; Shih 2002; Waldinger and Lichter 2003) and in Europe (Friberg 2012; Midtbøen 2014) demonstrates that employers use race and ethnic background as proxies of productivity, but that their views of minority applicants often are based on crude stereotypes. In this regard, England (1992) has made a useful distinction between statistical discrimination, on the one hand, and “error discrimination,” on the other, arguing that the latter refers to discriminatory practices guided by erroneous estimates of group averages, typically based on stereotypes about blacks or women. Importantly, however, the notion of error discrimination shares with statistical discrimination the view that employers do not necessarily have a general distaste against particular groups per se, but rather act in a discriminatory way “in an effort to hire a more productive workforce” (England 1992, 60).

3.1.3 Intergroup Relations

While discrimination is often theorized as part of decision-making processes at the individual level, collective phenomena such as stereotypes and prejudices, and their diffusion or change, are also part of the dynamics between individuals and groups. In everyday life, actors inevitably classify people into social categories where new information is assigned to existing categories. This categorization process is useful and even necessary to orient oneself in an environment rich in stimuli, information, and events. However, information confirming one’s own conviction tends to be stored, while those contradicting convictions tend to be disregarded, as it disrupts routine and means additional cognitive effort (Nickerson 1998). Categorization assigns individuals to social groups; it often entails the division of social space in an “in-group,” which includes the actor of categorization, as opposed to an “out-group.” Categorization relies on stereotyping, an inevitable by-product of normal cognitive processes. Stereotypes are “pictures in our heads,” according to the famous definitions by Lippmann (1922).

Through categorization, interpersonal behavior becomes intergroup behavior. An individual’s self-image results from both personal identity (i.e., what distinguishes one individual from all others) and social identity, the part of the self-concept derived from the consciousness of belonging to one or more groups. According to social identity theory, individuals look for a positive social identity (Tajfel and Turner 1979). As social identity is influenced by group membership, people tend to judge positively the group they belong to and compare it advantageously in relation to out-groups. The preference for the in-group improves the individual’s social identity; the identification with the in-group leads to favor it over out-groups, which is often called “in-group favoritism.” A group can maintain its higher status by giving privileged treatment to in-group members and reducing access to resources to out-groups. Experimental evidence shows that the simple fact of categorization may arouse intergroup tension between two groups of people randomly assigned to each group who share a common task (Tajfel et al. 1971).

While intergroup contact can lead to stereotyping, prejudice, and discrimination of the out-group, contact theory argues that it may also lead to decreasing prejudice and conflict between majority and minority group members. According to Allport (1954, 281), “[prejudice] may be reduced by equal status contact between majority and minority groups in the pursuit of common goals. The effect is greatly enhanced if this contact is sanctioned by institutional supports (i.e., by law, custom, or local atmosphere), and provided it is of a sort that leads to the perception of common interests and common humanity between members of the two groups). Against Allport’s assumption that ethnic antagonism is primarily “a product of fears of the imagination,” other authors identify the source of intergroup attitudes and conflict in functional relations between groups and their competition for scarce resources (Katz 1991). When the interests of the groups are interdependent, the group members are supportive and cooperative with each other; when the interests of one’s own group and the other groups are in conflict, competition arises. Negative attitudes toward out-groups originate from a feeling of threat (LeVine and Campbell 1972; Esses et al. 2005). Indeed, threat theory is a staple in research on attitudes to immigrants and their descendants.

Realistic conflict theory states that the higher the competition over limited resources, the higher the prejudice and the hostility between groups (Sherif 1966). Integrated threat theory extends the threat derived from the competition on tangible resources like safety, health, economy, and well-being, to the threat perceived on symbolic interests of the in-group, its beliefs, attitudes, and morals, thus echoing social identity theory (Stephan and Renfro 2002). Such threats may target the person or the whole group. A threat is a subjective perception: it does not need to be real. Such perception may, therefore, be constructed by media and public discourse (Brug et al. 2015). The attention to non-economic threats, such as identities, values, and beliefs, has enhanced the threat theory. According to this strand of the literature, the labor market considerations play a less significant role in shaping attitudes toward immigration when values and beliefs are accounted for (Hainmueller and Hopkins 2014).

How attitudes and behavior are linked is a much debated and controversial question. The assumption of a mechanical relationship, supposing that human action is the direct product of conscious mental states, is surely too simple and misleading. In a classical experiment, LaPiere (1934) documented that the articulation of racist attitudes does not need to convert in discriminatory treatment. The weak correspondence between explicit attitudes and behavior is confirmed in numerous studies (e.g., Pager and Quillian 2005; Blommaert et al. 2012). In contrast to the study of LaPiere, however, the disconnection goes more often in the direction of an apparent lack of prejudice and de facto discriminatory decisions. The affective dimension of prejudice (emotional prejudice) is found to be a better predictor of discriminatory behavior than cognitive dimensions (Talaska et al. 2008). The predictive validity of implicit associations as well as their link to discrimination outcomes are also a matter of controversy (Rooth 2010; Oswald et al. 2013; Dovidio et al. 2015; Carlsson and Agerström 2016; Bertrand and Duflo 2016).

In the seminal article “Attitudes vs. Actions”, sociologist Richard LaPiere showed that there is “no necessary correlation between speech and action” (1934, 231). The study took the form of an experiment where LaPiere traveled with a Chinese couple through the US in the 1930s, at a time of widespread bigot attitudes against “Orientals.” Only in one out of 251 instances did hotel managers refuse the couple accommodation. To provide a comparison between this (at the time) accommodating behavior and reported attitudes, LaPiere questioned 6 months later the same managers whether they would be willing to accommodate distinguished Chinese guests. Their response was overwhelmingly negative; only in one case, the answer was positive.

Brought together, individual theories seek an explanation for the phenomena of discrimination in the personal, internal motivations of perpetrators or in the processes assumed to be similar across countries and therefore universally valid (Guimond et al. 2014). Yet, as we saw, the association of motives and behavior is not straightforward. Underlining the difficulties of measuring motivations, Reskin (2003) recommends shifting the emphasis from individual beliefs and attitudes to the in-depth analysis of social mechanisms; that is, processes that mediate the link between internal states and discriminatory behavior. Many of such social mechanisms are found at the organizational level.

3.2 Organizational-Level Theories

Interpersonal and intergroup encounters always take place in socially structured contexts, making necessary an enlargement of scope to the meso-level of the organizational environment. Organizations – linking the micro and the macro social level – are key structural contexts shaping inequality (Baron and Bielby 1980). Mediating the impact of the individual-level mechanism of discrimination such as cognitive bias and stereotypes of the actors, organizational arrangements govern the extent to which ascriptive characteristics become relevant in determining social outcomes via the distribution of opportunities and rewards. An example in the labor market illustrates this mediating function. Organizational rules influence the degree to which recruiters are informed of ascriptive characteristics, which in turn influence selection-behavior. Facing incomplete information about candidates, recruiters interpret “signals,” notably of ascriptive nature, as decision-making tools. Blinding information is, therefore, a tool to curb the impact of unwanted bias. Studying recruitment of musicians in US orchestras, Goldin and Rouse (2000) demonstrated that the adoption of new organizational rules, here “blind auditions,” explained 30–50% of the increase of women among new hires. Organizational practices are shaped by societal mechanisms; as such, they might be seen as “the immediate causes of variation in ascriptive inequality” (Reskin 2003, 12).

Tilly (1998) emphasizes the importance of organizational dynamics in creating and maintaining group boundaries. Moreover, he develops an organizational account of “categorical inequalities” (i.e., inequalities across groups of people on the basis of rigid social categories such as gender, race, and immigrant status). According to Tilly, inequalities are not caused by attitudes and beliefs but by the organizational structures and the matching of the exterior (i.e., social) categorical distinctions, to interior organizational distinctions, such as jobs. Interior job distinctions are socially more powerful and generate larger inequalities when they overlap with exterior and culturally legitimate social categories. Distinctions between categories (e.g., men and women, white and black, citizens and non-citizens) are used to both distribute and legitimate inequality. Two complementary mechanisms are primarily responsible for inequalities across social categories: Exploitation, which amounts to unequal distribution of rewards proportionate to value produced, and opportunity hoarding, which amounts to excluding others from access to resources (e.g., jobs). The durability of inequalities depends on their organizational anchoring.

3.2.1 Organizational Procedures: Formalization

Studying the organizational determinants of recruitment has a long history in sociological research. In his famous theory of the modern bureaucracy, Weber (1946), for example, argues that formalized procedures constrain managerial discretion. Merton (1957), too, emphasizes how formal procedures in bureaucracies ensure control over effective decision-making. In the essay “Bureaucratic Structure and Personality,” he notes that “specific procedural devices foster objectivity and restrain the ‘quick passage of impulse into action’” (Merton 1957, 195).

Organizational and psychosocial theories indicate that the formalization of recruitment and promotion through bureaucratic practices is most likely to counter bias and discretionary decisions in access to employment, as they mediate the impact of individual-level mechanisms (Reskin 2000). Bielby makes this argument most clearly. He argues that “the impact of gender and racial stereotyping on judgments about individuals can be minimized when judgments are based on timely and relevant information; when decision makers evaluate that information consistently with respect to clearly articulated criteria; and when a mechanism exists for holding decision makers accountable for the process they have used and criteria they have applied in making their judgments” (Bielby 2000, 124). Following structural theorists of inequality, mainstream policy recommendations promote formalization of procedures as the proper organizational remedy to harness biased behavior.

However, analyses of observational data measuring the impact of bureaucratic approaches casts doubts on their overall efficacy, suggesting that some approaches being more effective than others (Sturm 2006; Kalev et al. 2006). Controlling managers’ discretion and bias proves counterproductive as it may stir resistance and may have adverse effects. In their studies aimed at assessing the effectiveness of antidiscrimination organizational policies, Dobbin and Kalev (2013) and Dobbin et al. (2015) identify the creation of formal organizational responsibility in charge of developing equal opportunity programs ensuring internal compliance to the regulatory frame as crucial tools to enhance the diversity of the workforce. Transparency of the allocation process and open accountability for the decisions proved also effective in increasing diversity.

3.2.2 Organizational Mechanisms: Networks as Opportunity Hoarding

Because of their mediating role, organizational structures may attenuate categorical distinctions – as with formalized procedures – or indeed accentuate them. This is the case when employees’ referrals are used largely in the recruitment process. While cost-effective and promising a better fit of newcomers in the workforce, this practice of activating internal networks, however, might prove to be a mechanism for ensuring in-group preference and promoting “homosocial reproduction” (Kanter 1977), whereby the dominant group favors and gives advantages to individuals carrying their ascriptive characteristics, in terms of ethnic background, racial appearance, and sex.

Resorting to networks to fill a position amounts to monopolization of resources by the established group to the detriment of “outsider” groups. Such referral practices result in the exclusion of categorically distinct others from jobs: as a mechanism of “opportunity hoarding” (Tilly 1998), it powerfully contributes to the reproduction of existing inequalities. Boulton (2015) provides an empirical example of this mechanism with his qualitative analysis of three large advertising agencies and their practices in the allocation of highly sought-after internships, which constitute a crucial point of entry into the labor market. Under the cover of color-blind meritocracy, influent players place friends and relatives, thus ensuring the material advantage of the established racial group.

As aptly noticed (Voss 2010), this mechanism is close to Weber’s idea of social closure. Networks are effective ways in not only gaining access to employment (as well as housing or services), but also in securing further education, informal mentoring, and other tools leading to career advancement. Although apparently neutral, activation of networks results in powerful instruments of cumulative (dis-)advantage.

3.2.3 Organizational Environment: The Regulatory Framework

Organizational practices are shaped by societal mechanisms. The mediating function of organizations derives also from the fact that they represent the implementing level of general policy orientations. Describing the history of corporate policies and tools in the US, Dobbin and Kalev (2013) illustrate how the macro regulatory frame was responsible for the implementation of antidiscrimination policies at the corporate level and influenced the way those policies evolved over time. In the UK, the institution of the Commission for Racial Equality in 1976, on the basis of the Race Relation Act, has progressively made the regulatory framework for businesses and public services more precise and stringent. In the EU, the 2000 Directive “implementing the principle of equal treatment between persons irrespective of racial or ethnic origin” (2000/43/CE) and the one “establishing a general framework for equal treatment in employment and occupation” (2000/78/CE) have similarly shaped the regulatory frame inspiring national legislation influencing organizational setups (see Chap. 7).

While the analysis of the regulatory frame has stimulated a vast literature, the impact of its enforcement is less developed. A crucial issue in this respect is how extensively and effectively the regulatory frame succeeds in preventing discrimination. Assuming that employers discriminate, consciously or unconsciously, as long as this is de facto possible, Petersen and Saporta (2004) shift their analysis to the conditions under which discriminatory practices in hiring, salary, promotion, or departures are more expected to occur. Analyzing the whole career development of employees in a large US service organization, they find that the hiring process appears as presenting the widest “opportunity structure for discrimination.” It is the most exposed to risks of discrimination because this is where the chance of employers being “caught in the act” is most limited (see also Bendick Jr. and Nunes 2012, 242–243).

While Petersen and Saporta analyze the room left uncovered by the regulatory frame, Hirsch (2009) focuses on the mechanisms ensuring efficacy to such a frame. Studying the direct impact and indirect pressure of legal and judicial enforcement of antidiscrimination legislation in the US, she shows that the case-by-case regulatory approach is not directly effective on the sanctioned discriminatory companies. Yet, sanctions exert an indirect pressure by creating a normative environment promoting gender and racial equality: “the driving force of the law is not sanctions but the legal environment they create” (Hirsch 2009, 245). However, gender desegregation has proven more sensitive to this normative pressure than race desegregation, as enforcement efforts in the latter respect lack sustained political support in comparison with those for sex desegregation (Hirsch 2009, 268). In the EU, the implementation of the directives at the corporate level is quite limited. With these insights in mind, it is not surprising that in their meta-analysis Zschirnt and Ruedin (2016) reported no difference in levels of hiring discrimination before and after the introduction of the EU directives.

Becoming aware of the mediating role of organizations has a bearing on the research agenda on discrimination: insights from social psychological research on prejudice and stereotypes are thus coupled with sociological research on the dynamics of organizations and institutions, providing analyses in which the organizational contexts of discrimination are moved to the forefront of this field of research. Yet, in turn, organizations are situated in larger social, economic, political, and legal environments exerting a powerful influence on the organizational settings.

3.3 Structural-Level Theories

Structural discrimination shifts the attention precisely toward such broader societal structures. The contextual dimension neglected in early theories (Fiske 1998) provides tools to understand variations in discrimination across time and space and the way it is produced and reproduced by institutions. Compared to individual and organizational theories, a structural discrimination approach expands the analysis of discrimination usually confined to one domain and a point in time in the two significant directions of time and scope (Pager and Shepherd 2008). Time, by emphasizing the production and reproduction of inequality into enduring self-perpetuating phenomenon through racial bias. Scope, by transcending unequal treatment in a specific domain, and paying attention to the interrelations among various domains affecting the entire society.

3.3.1 Present as Sediment of the Past

The advantages of one individual or group over another accumulates over time, reinforcing disparities so that the inequality of this advantage grows over time. Merton (1968) speaks in this regard of the “Matthew effect,” referring to the “parable of the talents” in the Book of Matthew. Cumulative advantage presents an affinity to theories of social stratification and reproduction linking social class origin to allocation mechanisms and social outcomes, reproducing the society class structure (Bourdieu and Passeron 1990). The cumulative advantage is the unequal growth rate in an outcome for individuals holding different statuses and growing inequality over time in a status group.

The cumulative disadvantage is the reverse side of the cumulative advantage (DiPrete and Eirich 2006). In its most frequent sense in sociology, the process of cumulative disadvantage is understood as the combination of direct and indirect effects of group membership on outcomes (negative for minority groups in relation to the majority) at different stages in the life course (Blank 2005). Cumulative disadvantage focuses on differential outcomes over time within a particular context, emphasizing dynamic processes that reinforce disparate outcomes. Blau and Duncan (1967) developed this concept in their classical study of The American Occupational Structure, yet it may easily be applied to similar cumulative disparities among ethnic groups.

When the timespan considered exceeds the lifespan to encompass generational succession, the attention shifts to history. Historical practices and policies of intentional discrimination project their gloomy shade into the present time through the mechanism of cumulative disparities. Therefore, historical experiences of exclusion may actualize disadvantage over time. This sort of structural discrimination is known as “past-in-present discrimination” (Williams 2000). Affirmative action policies were designed to counter the phenomenon’s inertia of this disadvantage (Wrench 2007).

In the US, the history of slavery and institutionalized racial segregation affects structures of disadvantage particularly concerning the African American population (Massey and Denton 1993; Pager and Shepherd 2008; Alexander and Rucker Jr. 2010). In many European countries, such as Britain, France, and the Netherlands, large migration inflows in the post-war era came from former colonies echoing the longstanding history of imperialism and colonialism (Castles et al. 2014).

Europe’s colonial past also has a bearing on contemporary patterns of racism: histories of exploitation directly affect ethnic relations through representations, ideologies, and practices that convey negative perceptions of minorities as inferiors and deny them full membership in the majority community (Bancel et al. 2010; Gilroy 1987, 2005; Oostindie 2008; Back and Solomos 2000; Thomas 2013; Amiraux and Simon 2006). Racism and anti-immigrant sentiment in Europe are also related to the economic and social consequences of the economic crisis in the 1970s and, later, to the focus on security and global terrorism following September 11, 2001. The combination of large-scale migration and a revival of nationalism and its symbols have created a situation that systematically works in disfavor of migrants in general, and of the Muslim population in Western countries, in particular (Castles et al. 2014).

3.3.2 Cumulative Interrelated Processes

Analyses of discrimination at the societal level expand further in a second direction by enlarging the scope of the analysis to cumulative processes. If cumulative disadvantage focuses on differential outcomes over time within a particular context, Blank goes beyond the dynamic progression with her concept of cumulative discrimination, defined as “discriminatory effects over time and across domains” (Blank 2005, 2; Blank et al. 2004).

Discrimination may indeed cumulate across processes within a domain of social life, such as the labor market: discrimination in hiring or work assignments, for instance, may affect promotion prospects and wage growth. Moreover, discrimination in one social domain may have spillover effects from one domain another. Consider as an example, the following sequence of effects: Discrimination in housing shapes residential patterns (Massey and Denton 1993). Such patterns, in turn, affect the concentration of minority students in schools, in traditional catchment area systems, where students are assigned to a public school depending on the geographical area in which they are domiciled. The combined impact of the socioeconomic and the ethnic composition of the school have an effect on student performance (Karsten 2010) and, in turn, unemployment risks (Heath and Cheung 2007; Heath et al. 2008). Furthermore, residential patterns have an impact of occupation: unemployment rates are higher when job opportunities are located far away from the neighborhoods where people live (spatial mismatch; Duguet et al. 2009; Kain 1968). Blank et al. (2004) thus enlarge the scope of the analysis to encompass the interrelations among different domains, stressing the systematic aspect of the cumulative process. However, acknowledging the difficulty of the task, scholars regret that research in this direction is rare.

Blank et al. (2004) theorize cumulative discrimination as disadvantages across time and domains combined with causal analysis. Reskin (2012) similarly embeds it in a “system perspective” with her notion of “über discrimination” (see Chap. 2). According to Reskin, sociologists have been too concerned with patterns of discrimination in particular social areas, preventing high-quality analyses from addressing the “reciprocal causality of disparities across spheres” (Reskin 2012, 18). The lack of a systems perspective on racial inequality in mainstream quantitative research renders invisible the potential feedback effects by which patterns of disadvantage are transferred across time and domains, and, as a result, prevents policy interventions from advancing racial justice. Reskin thus calls for increased attention to the relations among subsystems, of the feedback effects reinforcing disparities across subsystems, sustained by beliefs and values influencing the distribution of resources.

3.3.3 Institutional Discrimination as a Result of State Policies and Practices

If cumulative processes in time and scope build the core of structural discrimination, Pager and Shepherd (2008) subsume under this label also a somewhat different conceptualization, often labeled “institutional discrimination.” They define it as “the range of policies and practices that contribute to the systematic disadvantage of members of certain groups” (Pager and Shepherd 2008, 197), be they carried out by state or non-state institutions toward racialized or ethnicized groups.

Embedded in the radical black tradition that can be traced back to W. E. B. Du Bois, the theory of institutional racism was originally formulated by Carmichael and Hamilton (1967). In their analysis of the disadvantage of blacks in the US, they show no interest in intentions and interpersonal situations but focus on the effects of socially established power relations. Carmichael and Hamilton claim that racist practices are at the heart of ordinary practices; racism, therefore, finds its place in its daily banality, without the need for justification. In this perspective, racism is inherent in the very functioning of society, embedded in routine mechanisms ensuring the domination of certain groups. Because of its routinization, there is no need for any scientific theory or justification. Institutional racism is “less overt, far more subtle, and less identifiable in terms of specific individuals committing the acts. However, it is no less destructive of human life. It originates in the operation of established and respected forces in the society and thus receive less public condemnation” (Carmichael and Hamilton 1967, 20). In this perspective, racism is part of the very functioning of society, ensuring through routine mechanisms the domination of already privileged groups.

As pointed out in the previous chapter, the concept of “systemic racism” is very close to institutional racism. It refers “to the foundational, large-scale and inescapable hierarchical system of US racial oppression devised and maintained by whites and directed at people of color” (Feagin and Elias 2013, 936). Systemic racism is a “material, social, and ideological reality that is well embedded in major US institutions” (Feagin 2006, 2). However, grounded in the race-critical literature, it adds to it the notion of the white frame, “a socially constructed, meta-structure shaping and pervading not only the ‘state’ but also the ‘economy’ and ‘civil society’” (Feagin and Elias 2013, 937) permeating all aspects of US society. The white frame concept confers materiality and visibility to the actual majority, white promoters of systemic racism, otherwise hidden behind abstract references to “society.” Systemic racism here differs from the organizational systemic discrimination discussed at the organizational level, as it emphasizes societal-power unbalances and the role of perpetrators and perpetuators of racist practices as causes of inequalities.

This contextual, institutionally embedded dimension of discrimination has found high resonance in various European countries, with an increasing focus on discrimination in systems of equality (see Chap. 5). The concept of “institutionalized racism” made its way to the UK as applied to colonial immigration. It is prominently featured in the Macpherson Report of 1999, resulting from the judicial enquiry in the murder of a young black person in an unprovoked, racist attack and in the failure of the police investigations into this murder. The report presented to Parliament by the Secretary of State for the Home Department heavily criticizes those investigations as “marred by a combination of professional incompetence, institutional racism and a failure of leadership by senior officers” (Home Office 1999, para 6.34). The authors outline this concept as follows: “The collective failure of an organization to provide an appropriate and professional service to people because of their color, culture, or ethnic origin. It can be seen or detected in processes, attitudes, and behavior which amount to discrimination through unwitting prejudice, ignorance, thoughtlessness and racist stereotyping which disadvantage minority ethnic people” (Home Office 1999, para 6.34).

In Germany, the concept is adopted under the label of institutional discrimination (Gomolla 2017) to designate the production of inequalities by institutions intended to provide services equal services to all individuals (see Chap. 5). Gomolla and Radkte (2000) theorize institutional discrimination in their analyzing of school failure of children of immigrants. The core of the matter lies not in prejudice or the intention to discriminate of the parties involved, but in the durable and systematic nature of relative disadvantages produced by the school structure and functioning. Analyzing statistically measurable effects of the unequal distribution of educational success by ethnic differences, Gomolla and Radkte (2009) shift the attention away from the individual and the interactional levels toward the legal and political framework conditions, the organizational and financial aspects, the structures, programs, norms, rules, and routines as well as collective knowledge repertoires supporting decision-making. They focus on institutions in the Durkheimian sense, as a system of social relations with certain stability over time, with collective ways of acting and thinking and with their own existence outside individuals. From the perspective of institutional discrimination, critical questioning of existing institutions works as a programmatic tool and lays the foundations for the search for reforms and affirmative action policies aiming at justice and equity (Gomolla 2017).

As for France, “the existence of systemic racism within certain institutions (particularly the police, schools, social housing, and public health services) produces widespread discriminations and contributes to segregation” (Amiraux and Simon 2006, 206). Yet the development of the genuine sociology of ethnic minorities has been hindered by the French, republican integration model. This is both a political fact and a largely dominant “a-racial” (Amiraux and Simon 2006, 204) analytical referent, based on the principle of in-differentiation and assimilation. The heated debate about the use of ethnic and racial categories in statistics is symptomatic in this respect (Simon 2015).

Remarkably, scientific attention and political sensitivity to ascriptive inequality of immigrant-origin groups in Europe grow parallel to their long-term settlement in European immigration societies, revealed by the emergence of migrant offspring as a social reality and political actor. Yet in today’s immigration countries, group hierarchies are institutionally anchored in state policies and practices. Regulation of immigration increasingly diversifies status tracks, thus producing a “legal stratification of immigration status.” “Immigration and citizenship laws continue to create hierarchies among migrants that mirror the intersection of non-meritocratic attributes of social group membership such as gender, race/ethnicity nationality, religion and class” (Ellermann and Goenaga 2019, 2).

In the highly stratified political and economic international system of nation-states, the automatic acquisition of citizenship by birth determines critically unequal access to resources for individuals. In a provocative book, The Birthright Lottery, Shachar (2009) develops the analogy between birthright citizenship in rich societies and the inheritance of property, which opens access to rights and secures privileges. By virtue of this comparison, birthright citizenship amounts to an ascriptive attribute in the face of global inequalities. As such, it contributes to the production and reproduction of inequality into a self-perpetuating phenomenon nurturing processes of cumulative advantage. Making this point, Shachar shows us yet another example of structural discrimination.

3.4 Conclusion

This overview of main theories in the field shows the complexity of discrimination phenomena, reflecting such pervasive domination relationships that they materialize at every level of analysis of social behavior – individual, organizational, and structural. In spite of these different levels of analysis, the various theories of discrimination reviewed share a common feature, namely the fact that discrimination maintains privileges of certain ascribed groups over others. As such, discrimination helps to reproduce existing power relations among groups and consequently perpetuates ethnic and racial hierarchies. Perpetrators – consciously or not – make use of their power to engage in discrimination to uphold their privileges at the detriment of individuals and groups in a less favorable position in the social hierarchy.

For a long time in Europe, the dominant frame to understand social and economic inequality was social stratification without references to ethno-racial diversity. This interpretative frame was applied also to labor immigration after World War II. Yet the changing features of immigration (settlement of early migrant populations, development of migrant and refugee flows at a larger global scale, descendants of immigrants coming of age) combined with deep socioeconomic transformations of receiving societies have gradually uncovered how social hierarchies are intertwined with and overlap with ethnic and racial hierarchies. Indeed, Fassin and Fassin’s (2006), From the social question to the racial question?, is the evocative title of a stimulating essay pointing in this direction.

The multi-layered theoretical approaches show the importance of the macrosocial dimension. The European context is diverse by the number of countries yet similar to its normative frame lends itself for comparative studies aiming at highlighting the relevance of the structural and institutional dimensions shaping forms and scope of ascriptive inequality.