Barbara van der Ent (6056970) Thesis - Research Master Social Sciences 9 November 2015, University of Amsterdam Supervisor: Prof. Dr. Jan Willem Duyvendak Second reader: Dr. Sébastien Chauvin
Ethnic labour market discrimination and how ‘Moroccan‘ and
‘Surinamese-Hindustani’ Dutch can compensate for ethnic penalties
ABSTRACT
Recent research in The Hague (the Netherlands) has shown that ‘Moroccan’ and ‘Surinamese-Hindustani’ Dutch face ethnic labour market discrimination compared to ‘autochthonous’ Dutch. However, they both can counter this discrimination in different ways. ‘Moroccan’ Dutch can compensate for the ethnic penalty by having more work experience than ‘autochthonous’ Dutch. ‘Surinamese-Hindustani’ Dutch can compensate by stressing their integration into Dutch society (Andriessen et al., 2015). This article shows why both groups can compensate for the ethnic penalty differently, and explains when and how symbolic boundaries between social categories can be crossed. It will contribute to theory about boundary making and emphasize the role of emotions in boundary making processes. To reveal the stereotypes, expectations and emotions employers have, twenty semi-structured interviews were conducted with employers in the hospitality and retail sector. It is shown that discrimination of ‘Surinamese-Hindustani’ Dutch candidates is not caused by negative stereotypes of the group, but rather by the fact that their name is not recognized as Surinamese-Hindustani: since they are unknown, they are unloved. For ‘Moroccan’ Dutch candidates, negative stereotypes and expectations do play a role. This group is more negatively evaluated and different types of anxiety on part of the employers play a crucial role. For this group holds that: as they are known, they are unloved. In discrimination research, more attention should be paid to different stereotypes and emotions that underlie the different treatment of specific discriminated groups.
Keywords: labour market discrimination, ethnic penalty, stereotypes, boundary making,
compensation, anxiety
1
INTRODUCTION
Ethnic labour market discrimination is a fiercely debated topic. Dutch media regularly report on individual examples of job or internship discrimination based on ethnicity. The case of Jeffrey Koorndijk, who applied for an internship at a computer store, caused great controversy and emotional responses. The employee who dealt with his application accidentally replied to Jeffrey, instead of his employer, with the message: “It is nothing. First of all, he’s black (negro). Secondly, no experience with computers” (Volkskrant, 2014, my translation). Although the company said it was meant as a joke, it generated a lot of resentment.
Prime Minister Mark Rutte gave rise to controversy in March 2015 when he argued in an interview that discrimination in the Netherlands exists, but that paradoxically the solution is in the hands of the discriminated individual: “New arrivals always had to adjust themselves, and had always faced prejudices and discrimination. You have to fight yourself in” (Umar, 2015, my translation). Nevertheless, diminishing labour market discrimination is an important policy objective for the government. In 2014, a plan of action was released to reduce labour market discrimination which emphasized that every form of discrimination is unacceptable, because everyone deserves a fair chance and equal treatment (Ministry of SZW, 2014, p. 1).
A substantial gap exists between the unemployment rate of ‘autochthonous’1 Dutch and other citizens. In the Netherlands in 2014, the unemployment rate of first and second generation migrants with a ‘non-western’2 background was 16%, whereas it was 5% for ‘autochthonous’ Dutch people. For youth between 15 and 24 years old, the unemployment rate was respectively 28% and 10% (Huijnk et al., 2014). Andriessen et al. (2012a) have shown that individual characteristics such as age, work experience and level of education (i.e. differences in human capital) could only partly explain the difference in unemployment rates. Controlled for relevant factors and productive attributes, about half of the gap remains unexplained. The composition of a persons’ network possibly plays a role. If ‘allochthonous’ Dutch lack certain ties with persons holding positions of power, it is more difficult to get access to labour market positions (Rydgren, 2004, p. 711).
Another feasible explanation is that ethnic penalties play a role. Ethnic penalty means that ethnic minority members have a disadvantaged labour market position “due to ethnic/racial attributes that impose societal barriers over their labour market integration” (Gracia et al., 2015, p. 1). In other words, they suffer from discrimination. Their disadvantaged labour market position can be explained by the discriminatory practices of employers.
1 I put terms as ‘non-western’, ‘autochthonous’ and ‘allochthonous’ between quotation marks, because I find these terms problematic and often exclusionary. However, the use of these terms are common in society and social research and no adequate alternatives are available now. I will use the terms ‘Moroccan’ Dutch and ‘Surinamese-Hindustani’ Dutch to refer people who were born in the Netherlands but had parents that were born abroad (i.e. second-generation migrants). I will not use quotation marks if I write about the origin of the name.
2 Non-western refers to someone who is born in Turkey, Africa, Latin America or Asia (with the exception of Indonesia and Japan), or to someone whose parent(s) are born there (CBS, 2015).
2 Discrimination can be defined as the detrimental treatment of individuals based upon the social category (e.g. an ethnic group) they belong to or are ascribed to (Veenman, 2010, pp. 1806-1807). This article will focus on this latter explanation; discrimination, or the ethnic penalty regarding access to the labour market. Although statistical analysis of the unexplained unemployment gap cannot fully prove the existence of discrimination – the unexplained rest can be the result of not included but relevant variables such as network effects – correspondence tests have adequately shown that discrimination indeed plays a role (for other methods to investigate discrimination see: Dolfing & Van Tubergen, 2005; Veenman, 2010).
In correspondence tests, fictitious candidates apply for genuine vacancies or put their resume on a jobseeker site. The candidates are equal in terms of age, level of education, language proficiency, motivation and work experience. The only difference is their name, which is a signifier of their ethnic background. A significant difference in the probability to be invited for a job interview demonstrates that ethnic discrimination plays a role. Numerous researches that used correspondence tests internationally and in the Netherlands, have shown that discrimination occurs and that some groups have lower chances at the labour market than others (Bovenkerk et al. 1994; Bertrand & Mullainathan, 2004; Dolfing & Van Tubergen, 2005; Pager et al., 2009; Andriessen et al., 2010; Kaas & Manger, 2011; Andriessen et al., 2012a; Blommaert, 2013; Panteia, 2015).
Recently, Andriessen et al. (2015) determined the extent of ethnic labour market discrimination in The Hague (The Netherlands)3 with a similar research design. The authors showed that ‘Moroccan’ Dutch and ‘Surinamese-Hindustani’ Dutch face discrimination; they had significantly lower chances to be invited for a job interview than similar qualified ‘autochthonous’ Dutch. Furthermore, the research added that discrimination works differently for ‘Moroccan’ and ‘Surinamese-Hindustani’ Dutch by showing that both groups could compensate for the ethnic penalty by adding specific information to their applications. Two conditions – additional work experience and integration – were tested in the correspondence tests. The conclusion was that ‘Moroccan’ Dutch could increase their chances at the labour market if they had two more years of work experience, extra job trainings and a paragraph about their motivation and productivity in their letter. ‘Surinamese-Hindustani’ Dutch, on the other hand, could compensate by adding a paragraph about their integration in and identification with Dutch society. Additional work experience did not increase the chances of ‘Surinamese-Hindustani’ Dutch and the integration condition did not diminish ethnic penalties for the ‘Moroccan’ Dutch candidates.
The results of the correspondence tests of Andriessen et al. (2015) ask for further investigation. Whereas the authors show that discrimination exists and compensation for ethnicity is possible, the question remains why compensation works differently for both groups. The empirical question in this article is thus twofold. Firstly, I will show why ‘Moroccan’ and ‘Surinamese-Hindustani’ Dutch face labour market discrimination. Ethnic
3 stereotypes and corresponding expectations and emotions of employers will be discussed, based upon semi-structured interviews with employers. The second question is why compensation mechanisms work differently for both groups. In other words, how can the ‘Moroccan’ and ‘Surinamese-Hindustani’ Dutch applicant alter the stereotypes and expectations of employers, so that they increase their job chances?
This article aims to contribute to discrimination theory by combining classical explanations of discrimination by economists (e.g. Becker, 1957; Phelps, 1972; Arrow, 1998) and social psychological and sociological theories (e.g. Tajfel, 1982; Lamont & Molnár, 2002) about social categorization and social boundary making. It adds that the boundaries between social categories in the working place are permeable. A differentiation should be made between specific discriminated ethnic groups to enhance the understanding of discrimination and its possible solutions. Furthermore, this article will show that taking the emotion anxiety and its different forms into account, is of crucial importance to explain personnel selection.
This article begins with a short summary of Andriessen et al. (2015) since it was the starting point for this research. An understanding of the conditions that were used, is necessary to understand the compensation mechanisms. A theoretical framework about different discrimination theories and the central role of social categories and boundary making, follows this part. Unequal access to labour market resources is an example of a social boundary. Symbolic or categorical boundaries (Lamont & Molnar, 2002; Wimmer, 2013) underlie this unequal access and insights in these boundaries facilitates the comprehension of discrimination. Stereotyping and intergroup emotions will be discussed, as they are crucial in social categorization and its consequences (Mackie et al., 2008; Cottrell & Neuberg, 2005; Ray et al., 2014). After a methodology section, the empirical part will address the overall preference for ‘autochthonous’ Dutch and the specific stereotypes and emotions related to ‘Surinamese-Hindustani’ and ‘Moroccan’ Dutch of employers. Employers expect few difficulties when hiring the former, more stereotypes and emotions are invoked by Moroccan names. These differences help to interpret the different compensation mechanism and explain when symbolic boundaries can be crossed.
CORRESPONDENCE TESTS AND COMPENSATION MECHANISMS
To determine the extent of ethnic labour market discrimination in The Hague Andriessen et al. (2015) sent more than 900 digital applications to 325 vacancies. The research consisted of two parts. In the first part, three fictitious candidates applied to 176 vacancies with a comparable resume and letter of motivation. The only difference was their name; Dutch, Moroccan and Surinamese-Hindustani names4 were used. The research showed that candidates with a Dutch name had a significant higher chance to be invited for a job
4 The choice for this group is based upon their relatively large number of inhabitants in The Hague and their relatively high unemployment rates (Andriessen et al., 2015, p. 15).
4 interview than candidates with a Moroccan or Surinamese-Hindustani name, despite being comparably qualified and suitable for the job (see graphic 1). No significant difference was found between the ‘Moroccan’ and ‘Surinamese-Hindustani’ Dutch candidates (Andriessen et al., 2015, p. 32). This part was called the baseline condition.
In the second part fictitious candidates applied to 149 vacancies. However, the candidates were not similar anymore. The application of the Dutch-named candidate remained the same, but the candidate with a Moroccan or Surinamese-Hindustani name changed in two different ways; a work experience condition and integration condition was used. The work experience condition and integration condition were randomly assigned to the ‘Moroccan’ Dutch and ‘Surinamese-Hindustani’ Dutch candidate.
In the work experience condition, the ‘Moroccan’ or ‘Surinamese-Hindustani’ Dutch was two years older, had two years additional work experience and additional and relevant job trainings. Furthermore, a paragraph was added in the letter of motivation5:
It is important for me to do my job well. To perform as best as possible, I have attended and successfully completed several job trainings. I consider myself to be a hard worker, which is confirmed by my current employer and my colleagues. (Ibid., p. 52, my translation)
The results showed that a candidate with a Moroccan name now has the same chance to be invited as the candidate with a Dutch name. The graphic shows a small difference in chance (27% for ‘autochthonous’ Dutch and 25% for ‘Moroccan’ Dutch), but this difference is not significant. The candidate with the Surinamese-Hindustani name, however, still has a significant lower chance to be invited (see graphic 1) (Ibid.).
The integration condition contains an extra paragraph wherein the ‘Moroccan’ or ‘Surinamese-Hindustani’ candidate wrote that he/she identifies as Dutch, feels at home in the Netherlands, is motivated to participate in Dutch society and is volunteering in a Dutch association. The paragraph refers to a possibly perceived social and cultural distance between the candidate and ‘autochthonous’ Dutch and tries to decrease this distance. For ‘Moroccan’ Dutch candidates a sentence about their religion was added, because the Islam is often portrayed as a dividing line that increases the experienced social and cultural distance (Ibid., 51). Furthermore, the candidate explained in the letter of motivation that people sometimes had questions about their ethnic background and mentioned the origin of their parents.
Finally, a personal note. In the past I have noticed that people have questions about my background although they do not always dare to ask. As you perhaps can see by my name, I am a Dutch person with Moroccan parents. I am Muslim, but in my opinion religion
5 We decided to add something to the letter and resume, because employers do not always read both. By adding in both documents, it is more likely that employers see the differences between the candidates.
5 belongs to the private sphere and not to the workplace6. I feel in particular Dutch and I
am motivated to participate in the Dutch society. My engagement with Dutch society is also apparent from the voluntary work I do for the National Foundation for the Elderly. (Ibid., p. 52)
The results showed that within this condition ‘Surinamese-Hindustani’ Dutch people have the same chance to be invited for a job interview as the ‘autochthonous’ Dutch candidate. The difference between ‘autochthonous’ and ‘Surinamese-Hindustani’ Dutch’ in chance to be invited is no longer significant. ‘Moroccan’ Dutch however, have a significant lower chance to be invited and thus still face discrimination (see graphic 1) (Ibid.).
Graphic 1: Chances of applicants to be invited in percentages (Andriessen et al. 2015).
To summarize, we have seen that ‘Moroccan’ and ‘Surinamese-Hindustani’ Dutch face ethnic labour market discrimination in the baseline condition, because they had lower chances to be invited for a job interview although they were similar to ‘autochthonous’ Dutch. Furthermore, Andriessen et al. (2015) showed that ‘Moroccan’ Dutch could increase their chances in the work experience condition, whereas ‘Surinamese-Hindustani’ could compensate for the ethnic penalty in the integration condition. Puzzling is why on the one hand the work experience conditions is effective for ‘Moroccan’ Dutch, but not for the ‘Surinamese-Hindustani’ Dutch, and on the other hand why the integration condition have effect for ‘Surinamese-Hindustani’ Dutch and not for the ‘Moroccan’ Dutch. The results indicate that both groups face discrimination (partly) for other reasons, because they need to act differently to increase their chances. To understand where discrimination comes from, we
6 For ‘Surinamese-Hindustani’ Dutch this sentence was absent.
34 27 26 19 25 14 23 17 24 0 5 10 15 20 25 30 35 40
Baseline Condition: work experience Condition: integration
6 will now turn to the theoretical section in which several discrimination theories and social categorization or boundary making are explained.
THEORETICAL FRAMEWORK
In discrimination research, two theories explaining why discrimination occurs are predominantly used; taste-based discrimination theory and statistical discrimination theory. In practice, however, it is hard to distinguish between them. Becker (1957) is the founding father of taste-based discrimination theory. He states “[i]ndividuals are assumed to act as if they have “tastes for discrimination,” and these tastes are the most important immediate cause of actual discrimination” (Becker, 1957, p. 122). Employers might, for example, dislike an employee with a ‘non-Dutch’ ethnic background and could therefore be reluctant to hire the person, unless they are willing to work for a lower wage (Ibid., p. 32; Andriessen et al. 2012b, p. 239). Taste can also be attributed to other employees or customers; when employers expect that their employees or customers will dislike certain people, they will avoid to hire them.
Whereas Becker (1957) emphasized that taste is the primary incentive to discriminate, statistical discrimination theory departs from market-based considerations. This theory implies that employers primarily want to maximize their profit, but do not have enough time and resources to be fully informed about the best suitable candidate (Phelps, 1972; Arrow, 1998). “Given the impossibility of measuring individual productivities in advance, employers may discriminate against whole classes of people based on (correct or incorrect) perceptions of the average productivity (or variation in productivity) of these classes” (Moss & Tilly, 2001, p. 86). Employers may use stereotypes about the group a candidate belongs to or is ascribed to (e.g. ethnic group, gender, religion), to estimate who is the most productive and reliable candidate. For example, some groups are overrepresented in criminality statistics or imprisonment and consequently individual group members can be perceived as ‘higher risk’ and less appropriate for a job. In other words, in-group variation is neglected, so all individuals within the stereotyped group are unjustly treated (Rydgren, 2004, p. 708).
Although taste-based discrimination and statistical discrimination differ in their primary source that explains unequal treatment – respectively taste and market-based considerations – it is hard to empirically distinguish the one from the other. A taste-based argument can be presented or legitimized in an interview as a consideration based upon statistics. Respondents can, for reasons of social desirability, decide not to show their overt taste against a group.
Besides reasons of social desirability, Gaertner and Dovidio (1986, used by Moss & Tilly 2001, p. 88) describe how overall the nature of racial prejudices and racism has shifted over the last decades, from overt to a more covert form: “[a]s more overt forms of racism become less common, racism more often takes an “aversive” form, entailing subconscious discomfort or anxiety about contact with members of another group, rather than conscious hostility”. Consequently, it can be expected that employers do not show their taste or preferences
7 openly, although they perhaps do hold these, and that they talk about it in more covert terms as felt anxiety and discomfort.
Crucial in understanding discrimination is social categorization, because the starting point of discrimination – whether it is based on taste or stereotypes – is that an employer categorizes people differently. The question is what the basis for this differentiation is. When and why are some people assigned to the in-group, whereas others are considered to be a member of the out-group? (Tajfel, 1982). As boundaries between social groups or categories are socially constructed or drawn, understanding why these boundaries exist and what their consequences are, is important when explaining discrimination.
Two kinds of boundaries can be distinguished; symbolic and social boundaries. Symbolic boundaries can be seen as categorical boundaries. These are conceptual distinctions people use to categorize, i.e. people and practices. Symbolic boundaries separates people into groups and causes feelings of group membership and solidarity (Lamont & Molnár, 2002, p. 168). A symbolic boundary makes clear who is in ‘your’ group (in-group) and implies inclusion of the people to whom you are similar. However, inclusion is intrinsically linked to exclusion. The people who are not on your side of the boundary are excluded (out-group).
Whereas symbolic boundaries are categorical boundaries, social boundaries can be defined as behavioural boundaries. They imply that people are treated differently because they are ascribed to another category. Wimmer (2013, p. 3) explains the relation between social and categorical boundaries clearly in his book ‘ethnic boundary making’:
Each identification (“I am a Swiss”) obviously implies a categorical boundary (the non-Swiss); each corresponding action (e.g. helping another Swiss to find an apartment in Los Angeles) implies discrimination against those on the other side of the divide (e.g. not helping someone from Sweden).
Social and symbolic boundaries are thus conceptually separate, but in reality intertwined. They “emerge when actors distinguish between different ethnic categories and when they treat members of such categories differently” (Wimmer, 2013, p. 3). Andriessen et al. (2015) have shown that social boundaries exist on the labour market in The Hague; equal qualified candidates are treated differently as a result of their name. The question however is, how is this related to the symbolic boundaries? Why and when are the categorical boundaries drawn and what are the implications? And more importantly, how can a symbolic boundary between ethnic groups be crossed to increase access to labour market positions?
Stereotyping plays an important role in social categorization or boundary drawing. Stereotypes are cognitive constructs about “traits that come to mind quickly when we think about groups” (Stanger, 2009, p. 2). These traits are seen as characteristic of a specific social group and, more specifically, of what differentiates this group from other groups. People are often unaware of the process of stereotyping and employers often make decisions based upon stereotypical thinking without being aware of it (Rydgren, 2004, p. 708; Dagevos, 2015). For the current research it is not relevant whether stereotypes are (to some extent) true, because
8 the crucial point is whether employers perceive them – conscious or unconscious – as true. Following the (slightly adjusted) Thomas theorem (Thomas & Thomas, 1928): If employers define stereotypes as real, they are real in their consequences.
Tajfel (1982) argued that people are inclined to relatively favour people in their in-group, while the people in the out-group are more negatively evaluated. Elias and Scotson (1965, p. xix) shed a light on this figuration of the established and the outsiders – in this research ‘autochthonous’ Dutch and ‘allochthonous’ Dutch – and show how stereotypical thoughts are formed:
[A]n established group tends to attribute to its outsider group as a whole the “bad” characteristics of that group’s “worst” section – of its anomic minority. In contrast, the self-image of the established group tends to be modelled on its exemplary, most “nomic” or norm-setting section, on the minority of its “best” members.
Siebers and Dennissen (2015) have argued that the Dutch discourse around migrants can be defined as cultural essentialism or cultural fundamentalism7. This discourse highlights the religious and cultural differences between particularly Muslim ‘migrant’ groups and ‘autochthonous’ Dutch. ‘Migrants’ are framed as “cultural subjects, bearers of cultural characteristics that are assumedly incommensurable, incompatible or contradictory to assumed cultural traits of non-migrant citizens” (Ibid., p. 474). This cultural incompatibility between ‘migrants’ and the ‘autochthonous’ Dutch is considered to be a problem of ‘allochthonous’ Dutch – because the Dutch ‘culture’ is presented as superior – and justifies forms of exclusion. In line with this discourse, it is expected that perceived or expected cultural incompatibilities play a role in discriminatory practices.
Social categories are cognitive constructs that are loaded with emotions. Turner and Stets (2005, p. 1) have argued that these emotions are crucial in understanding social relations. They emphasize that emotions are the glue that binds people together, generate commitment to groups and the larger society with its social structures, but also can drive people apart. Mackie et al. (2008) developed the Intergroup Emotion Theory (IET) which similarly claims that interactions between groups cannot be understood without taking into account the emotions people feel towards their own, and the other group (Mackie et al., 2008, p. 285). They contributed to the theory of social categorization by Tajfel (1982) by adding emotion as a factor, and clarify that IET “moves beyond the idea of a simple positive evaluation of in-groups and negative evaluation of out-groups, to focus on the distinct and differentiated emotional reactions that both in-groups and out-groups provoke” (Mackie et al., 2008, p. 1876).
Research has often not taken into account the diversity of emotions people have towards different groups (Cottrell & Neuburg, 2005, p. 770, Fiske, 2012). Including concrete and specific emotions related to groups, can contribute to our understanding of the
7 Some critical race scholars call this cultural racism. See for a discussion about the differences between those concepts Siebers and Dennissen (2015).
9 intergroup relations and behavioral responses (e.g. discrimination). Anger towards an out-group, for example, motivates confrontations with them. The feeling of fear, on the other hand, is more related to avoiding and moving away from the group (Cottrell & Neuburg, 2005, p. 773).
Furthermore, emotions are considered to be more important in predicting discriminatory behaviour and attitudes than beliefs. Talaska et al. (2008) have shown that emotional prejudices, which they defined as differentiated emotions towards out-groups, are two times better in predicting discrimination than stereotypes, which are beliefs about out-group characteristics. It is claimed that emotions are often seen as primary cause of behaviour in general, but this is particularly true for discrimination. “[S]tereotypes, beliefs and emotional prejudice, [are] all closely related to what people say they did or will do toward outgroup members, but emotional prejudices are more closely related to what people actually do” (Ibid., 2008, p. 284).
This theory section has shown that to understand unequal access to labour market positions (a social boundary), it is important to understand the corresponding symbolic boundaries. Stereotyping plays a crucial role in the process of boundary drawing and social categorization. The empirical focus in this article is on the one hand on the ethnic stereotypes and corresponding expectations and emotions of employers. In other words; what is the basis of the symbolic boundaries that are drawn? On the other hand, it will elaborate upon when these boundaries can be crossed and compensation for the ethnic penalty is possible.
METHODS, RESPONDENTS AND INTERVIEW TECHNIQUES
This article draws upon twenty semi-structured interviews with employers or employees who are responsible for personnel selection,8 conducted from April to June 2015. Whereas correspondence testing is a good method to reveal the actual amount of discrimination, it cannot show the intentions or motivations behind the discriminatory behaviour. Interviewing, on the other hand, can shed a light on the attitudes of employers. However, what employers say does often not predict what they do (Pager & Quillian, 2005; Veenman, 2010, p. 1813). The mixed methods design in this paper combines the results of correspondence tests with qualitative interview data in order to explain mechanisms to compensate for the ethnic penalty. Interview data will thus not be used to determine the extent of discrimination, but to deepen our understanding of why groups are discriminated against and why, when having added something to their application, they are not.
All interview respondents worked in the region of The Hague in the hospitality or retail sector. These sectors were large enough to find respondents willing to participate, while they were not included in the correspondence tests. The latter being a criterion, because I expected employers to be less willing to talk when they also had participated in the
10 experiments. Besides, I considered it important that interviewees were not only responsible for personnel selection, but also had experience in working together with their employees. For large companies in other sectors, personnel is often selected at HRM-departments without direct on-the-job experience, which was not the case for the respondents in this research. Furthermore, a specific characteristic of these sectors is that employees often have customer contact (except for cooks), which could increase customer discrimination (Pager et al., 2009, p. 15; Andriessen et al., 2010, p. 16).
Respondents were selected with a purposive sampling approach (Bryman, 2008, p. 458), using two websites with an overview of cafes, restaurants and shops in The Hague. At first, I randomly selected enterprises from lists on the websites. Later on, I searched more focussed to ensure a spread throughout the city district and to include different types of enterprises. Potential respondents received a letter with an interview request. This letter emphasized the expertise of the employers and their gatekeeper position. The research topic was framed as ‘selection processes at the labour market’. To avoid selectivity in the sample, the letter did not mention the research focus on candidates with different ethnic backgrounds and discrimination. Employers who did not want to participate mainly expressed a lack of time as most important reason.
Twenty interviews were conducted, equally divided from the hospitality sector and the retail sector. Seven of the interviewees were entrepreneurs, the other thirteen were on payroll. Eleven women and nine men were interviewed. Three of the respondents mentioned they were sometimes seen as ‘allochthonous’ or ‘non-Dutch’, because they had a ‘non-Dutch’ appearance. The interviews covered different types of hospitality (café, lunchroom, restaurant, cheap versus more expensive) and different shops (clothing stores, shoe stores, drugstore, supermarkets and construction markets). The interviews lasted approximately one hour, the shortest taking 41 minutes and the longest 1,5 hour.
The first part of the interview consisted of general questions about personnel selection, application procedures and the employees of the enterprise. The second part questioned the advantages and disadvantages of ethnic diversity in the enterprise. In the third part, vignettes were shown to the respondents. These vignettes were similar to the fictitious candidates in the correspondence tests (Andriessen et al., 2015) and consisted of cards with a name9, year of birth, place of birth, level of education and years of work experience. I asked the employers which candidate they might invite and whether they recognized the origin of the candidates name. In the first round of showing vignettes, all three candidates were equal. In the second and third round, the integration and work experience condition were shown by adding the same paragraphs that were used in the correspondence tests to the vignettes. All interviews were transcribed and coded. The coding was relatively open at first, as I searched the data for emerging themes. Based upon these
9 I used four different sets with names: 1) Michelle de Vries, Shaila Sietaram, Hafida el Kaddouri 2) Tessa Smit, Sunita Soekhoe, Bouchra Benaissa 3) Kevin de Bruin, Arun Sital, Mohamed Alaoui 4) Sander van Dijk, Narinder Ramcharan, Ahmed el Massaoui.
11 codes, I made a codebook and the coding process became more focussed on specific themes that turned out to be relevant (Bryman, 2008, p. 543).
Ethnic discrimination is a sensitive topic – most people do not want to be seen as having racist ideas and discrimination is prohibited by law – so social desirability in interviews can be a problem. Most employers choose their words carefully when explaining their opinion about ethnic diversity. Vignettes that present concrete scenarios (e.g. three different candidates that apply for a job) are useful to discuss sensitive topics (Ibid., p. 247) and I noticed that respondents began to open up in this part of the interview. Respondents were asked firstly who they thought the average employer in the Netherlands would invite for a job interview and why. Because the question was about other people, it was less threatening for the respondents (Ibid.) and made eventual discriminatory attitudes more acceptable to convey. Secondly, I asked employers which candidate they themselves preferred to invite and why. Emotions turned out to be important in personnel selection and in stereotypes about ethnic groups. It is important to note that when I write about emotions, I talk about representations of emotions by the employers. I did not neurologically measure their feelings, but rely on interview data and thus on the emotions presented to me.
It is worth mentioning that I am an ‘autochthonous’ Dutch woman, who also is perceived as such by others. One respondent stated that she would not be as open, should an ‘allochthonous’ interviewer have interviewed her. Although I sometimes encountered opinions that contradict my own, I tried to behave and react as neutral as possible.
EMPIRICAL ANALYSIS
This part is built up as follows. First, I will demonstrate that employers see personnel selection mainly as an emotional activity. Moreover, the lack of predetermined objective selection criteria may influence why some groups have lower chances at the labour market. Second, the preference for ‘autochthonous’ Dutch which employers show, is explained. Subsequently, I will elaborate upon the ‘Surinamese-Hindustani’ Dutch and show how employers perceive these candidates. The compensation mechanisms for ‘Surinamese-Hindustani’ Dutch are discussed and explained. Thereafter, stereotypes and emotions related to ‘Moroccan’ Dutch are elaborated upon and related to ways they can compensate for the ethnic penalty. The empirical analysis will end with the perceived grey area between smart entrepreneurship and discrimination.
Emotional selection: first impression and ‘the click’
When judging statements of motivation and resumes, employers mainly look at language proficiency and motivation. It is seen as positive if an applicant has knowledge about the specific enterprise and shows motivation for the particular place. Not every employer
12 demands a letter of motivation and resume. Some recruit new personnel via their network or just have persons drop by. In job interviews, employers mainly use two criteria. Firstly, they stress the importance of feeling a ‘click’ with the applicant. This is highly subjective and employers mostly refer to their feelings. Several employers explained that they often know in the first minutes whether they will hire a candidate or not. They use their intuition and the questions they ask in a job interview, are often not really important. The first impression is crucial and more experienced employers indicate they rely more on their feelings than in the beginning.
I first want to see a face, and then I know ‘yes or no’. It is a sort of feeling that you have with someone. [And where does that comes from?] Well, it’s a feeling. (Retail, R17)
Another criterion, which is strongly related to the ‘click’, is whether someone would fit into the team. Employers assess how likely a match with the team is to occur. Again, this is a subjective and emotional process; employers stress they intuitively feel whether a candidate will fit in, based upon the click. In most cases, a candidate is asked for a one day trial. This is a more objective selection criterion, because the decision to hire someone is based upon on the cooperation with the team. However, the invitation for a trial day is still based on the emotions of the employer.
I pay attention to the availability of the candidate and if it is a stable person. Those things are important for us. And you examine if someone would fit into the team, you can just feel it. You just feel if someone is a match with the team if you have a talk with that person. (hospitality, R9)
I can teach you to walk with a tray in 10 minutes, that’s not so difficult. But I cannot teach you the rest of it. How you function in a team, that’s impossible to learn. You need to have it already. And I need to have that feeling with you. (hospitality, R3)
To avoid discrimination, it is advised to work with predetermined selection criteria. This should make the selection process more objective and promotes equal chances for each candidate (Bielby, 2000). The expectation is that otherwise, an employer will often choose someone who is perceived as similar to themselves (in-group), thereby disadvantaging persons that are perceived as member of another social category (out-group). Employers often had, besides their intuition, also more objective criteria. They looked for someone with social skills, a tidy appearance (although it is debatable whether this is objective), language proficiency and flexible availability. Working experience was in these sectors not a decisive criterion.
None of the employers exclusively worked with a full list of selection criteria, all used their intuition and feelings to some extent. Some doubted the usefulness of a beforehand determined list with requirements. For example, one employer explained that he had hired
13 excellent employees who would never satisfy the objective criteria, but he had a good ‘click’ with them and it turned out to be a success.
Preference for ‘autochthonous’ Dutch candidates
Respondents were shown vignettes with a Dutch, Moroccan and Surinamese-Hindustani name. All respondents expected that the average employer in the Netherlands would choose the Dutch-named candidate first. Most respondents assigned the ‘Surinamese-Hindustani’ Dutch candidate to the second place, some argued that ‘autochthonous’ Dutch and ‘Surinamese-Hindustani’ Dutch will be assessed equally. They expected them to have equal chances on the labour market. ‘Moroccan’ Dutch candidates were seen as the group with the lowest chance to be invited by the average employer.
Reasons to prefer an ‘autochthonous’ Dutch candidate are centred around the expectation of similarity between the employer and the employee. A Dutch-named candidate is often expected to be alike in terms of their attitude to work, behaviour and culture. Choosing this candidate is seen as the safest choice, because an employer ‘knows’ what he or she can expect. Respondents described it as follows:
I think employers often would choose for the safest choice [= ‘autochthonous’ Dutch]. The one of which we know, if we hire this person, we will have profit of it without a lot of difficulties. (retail, R15)
You just choose for yourself the most recognizable candidate. You will choose that person. I don’t think it has to do a lot with expected motivation of the candidate, but the Dutch name is just closer to my name than Mohamed is. Yes. […] You will choose always the person that is the closest to you, so to speak. As you have to work together, it is pleasant if someone has the same background as you do. (hospitality, R1)
Well, maybe it is, what the peasant does not know, he will not eat [Dutch proverb]. Or they think that it will be easier [with the ‘autochthonous’ Dutch]. Or maybe they have the idea they know better what they can expect. Or they are afraid of cultural difference at the work place. […] [Are there also stereotypes about Dutch-names employees?] I don’t think that employers think about them in stereotypes. But more: “Okay, this is familiar, so it should be safer, or better or more easy”. While this is of course not true per se.
(hospitality, R6)
Employers could not always pinpoint specific stereotypes and expectations related to a ‘Moroccan’ or ‘Surinamese-Hindustani’ Dutch candidates. However, in emphasizing the importance of recognizability and similarity in relation to the ‘autochthonous’ Dutch candidate, they did show the expectation that ‘Moroccan’ or ‘Surinamese-Hindustani’ Dutch are different. They are seen as ‘others’, as a ‘surprise’ or as ‘less close’, thus as other social
14 categories. A symbolic boundary is drawn between them and ‘autochthonous’ Dutch. The relation between the symbolic boundary and the social boundary of unequal labour market chances is clearly demonstrated in the above quotations. A mild form of anxiety for or a feeling of uncertainty towards the ‘other’ or the ‘unknown’ underlies the decision to choose the ‘autochthonous’ Dutch candidate and disadvantage the other candidates.
Some employers emphasized that the preference for an ‘autochthonous’ Dutch candidate only exists when the employer is ‘autochthonous’ Dutch as well, or when the business is ‘Dutch’. “If I had a roti shop, I would choose for Arun. And if I sell kebab, I would invite Mohamed”, one employer (hospitality, R1) explains. They reversed the vignettes and claimed that the ‘autochthonous’ Dutch candidate would have no chance as well when applying to a restaurant with a Moroccan signature. It is claimed, in other words, that ‘allochthonous’ Dutch might avoid to hire ‘autochthonous’ Dutch and thus prefer members of other social categories. It is an empirical question whether this assumption is true.
Three respondents I interviewed considered themselves or were considered by others as ‘allochthonous’. They all had a light brown coloured skin, their parents were born abroad and they were born and raised in the Netherlands. They did not run a business with an ethnic signature. All three perceived variety in ethnic backgrounds and cultural differences as an enrichment of their business and their team. One of them explained he had no ‘autochthonous’ Dutch employees on payroll, because ‘allochthonous’ employees are more motivated, function better as part of a team and are less cheeky. Ethnic variety for him means thus only non-ethnic Dutch people. Two others were positive about both ‘autochthonous’ and ‘allochthonous’ Dutch and stressed the advantages of diversity and cultural differences. One of them suggested, and again it is an empirical question whether it is correct, that
If you have yourself another cultural background, you will not make this distinction so quickly. [Why do you think so?] Well, I think… If you have yourself a Turkish background or so, you cannot reject someone because he has a foreign name, because you have one yourself as well. Thus the stereotype will be gone. (retail, R19)
‘Surinamese-Hindustani’ Dutch candidates
To understand why ‘Surinamese-Hindustani’ Dutch candidates face discrimination and when they can compensate for the ethnic penalty to increase their chances, the first step is to map the stereotypes about this group that employers expressed. It can be expected that discrimination is the result of tastes and preferences (Becker, 1957) or is explained by the theory of statistical discrimination. In the latter theory employers draw upon stereotypes of the social group a candidate belongs to or is ascribed to, to estimate the productivity of the candidate (Phelps, 1972; Arrow, 1998).
However, few stereotypes came up for the ‘Surinamese-Hindustani’ Dutch candidates. Several employers did not know any negative stereotype at all. They explained that this group is not negatively portrayed by Dutch media and does not have a bad
15 reputation. One respondent (retail, R7) states: “Hindustani are a kind of neater and tidier people. And I think, yes, there is not bad reputation for them”. The few negative stereotypes about ‘Surinamese-Hindustani’ Dutch that were mentioned were not shared by more than two employers. Amongst others, they came up with the idea that these candidates are sometimes unreliable in keeping their agreements, are too family-minded (and this could contradict with responsibilities at work) or were only applying for jobs with status (instead of low-status jobs in the kitchen). Furthermore, two employers referred to the expectation that ‘Surinamese-Hindustani’ Dutch candidates can be too late or slow-workers.
On the other hand, employers mentioned several positive stereotypes about mainly Hindustani’ Dutch women and showed positive emotions. ‘Surinamese-Hindustani’ women are seen as social colleagues who can contribute positively to the atmosphere in a team. As we saw earlier, fitting into the team is an important criterion for many employers and for ‘Surinamese-Hindustani’ candidates the expectation is that it will be no problem.
‘Surinamese-Hindustani’ Dutch were perceived as closer to Dutch culture than ‘Moroccan’ Dutch. One employer (retail, R7) even stressed that “Surinamese people, I think yes, they are Dutch too. Because Suriname was also part of the Netherlands”. They are expected to have more or less the same work ethic and their behaviour is often seen as comparable to ‘autochthonous’ Dutch employees. Furthermore, for ‘Surinamese-Hindustani’ Dutch, the Dutch language is not expected to be problematic. Choenni (2011, pp. 18-19) shows that ‘Surinamese-Hindustani’ are indeed successful in terms of cultural integration. In general, they endorse Dutch norms and values about freedom of speech, freedom of religion and equality between men and women. Specific ‘Surinamese-Hindustani’ cultural values do not conflict with the cultural values of the majority in the Netherlands, so cultural conflicts barely occur. Choenni (Ibid., p. 18) also emphasized that ‘Surinamese-Hindustani’ seem to integrate successfully in socioeconomic terms and observes progress on the labour market.
Although progress is made, Andriessen et al. (2015) showed that in the baseline condition ‘Surinamese-Hindustani’ are discriminated against. ‘Surinamese-Hindustani’ Dutch and ‘autochthonous’ Dutch do have an equal chance to be invited for a job interview in the integration condition, in which they add voluntary work and personal information about the ethnic origin of their parents and their own identification with Dutch society. It was suggested that discrimination in the baseline condition is caused by the fact that employers do not recognize the name as being Surinamese-Hindustani. As soon as they can assign the candidate to the social category ‘Surinamese-Hindustani’ Dutch, discrimination disappears (Ibid., p. 57).
The interviews with employers show this is a reasonable explanation. Few negative stereotypes exist about ‘Surinamese-Hindustani’ Dutch and the ones I found were not broadly shared by others. None of the employers showed a taste against this group. There is no widespread image of ‘Surinamese-Hindustani’ Dutch as culturally different or poorly integrated in Dutch society. The sentences in which candidates show their engagement with Dutch society and their voluntary work are thus not the sentences that made a difference.
16 Furthermore, it was evident that employers had more difficulty with recognize e.g. Shaila or Narinder as Surinamese-Hindustani names, than in identifying the Moroccan names.
According to statistical discrimination theory, employers might rely on stereotypes to make their choice. In the case of applications with a Surinamese-Hindustani name, however, employers did not know to which social category the candidate could be ascribed and on which stereotypes they could rely. Employers did recognized the Surinamese-Hindustani name as ‘foreign’, but did not linked it to (positive) stereotypes about ‘Surinamese-Hindustani’ Dutch. The Dutch proverb unknown makes unloved (in Dutch: onbekend maakt onbemind) can be well applied to this group; a Surinamese-Hindustani name is often not recognized as Surinamese-Hindustani and the expectations are unknown, and consequently the person is ‘unloved’. When the candidate disclose their ethnic background, the symbolic boundary can be crossed and discrimination disappears. When they are known, they are loved. ‘Surinamese-Hindustani’ Dutch could not compensate for the ethnic penalty in the work experience condition. Even if they were two years older and more experienced, they still had a significantly smaller chance to be invited than the ‘autochthonous’ Dutch and the ‘Moroccan’ Dutch. Discrimination did not disappeared, because the reason for their discrimination is not related to specific doubts about their productivity or appropriateness. ‘Surinamese-Hindustani’ Dutch face discrimination because they are seen as the unknown other. They remain unknown in the work experience condition. It is remarkable, however, that even if the candidates have evidence in their resume of being successful in their last job (with six year work experience in the same company), their expected ‘otherness’ is so important that they will not be invited.
‘Moroccan’ Dutch candidates
‘Moroccan’ Dutch candidates are more negatively evaluated than ‘Surinamese-Hindustani’ Dutch. Employers often expect that the average employer would prefer to avoid hiring the ‘Moroccan’ Dutch candidate. More stereotypes about this group exist and they are more widely shared. Cultural and religious differences, criminality and a reluctance to hire a woman with a headscarf were themes that came up frequently. Moreover, employers reacted more emotionally on ‘Moroccan’ Dutch candidates. They explicitly talked about their emotions or sentiments in society – mainly about anxiety – or emotions were showed through their way of talking. More specifically, different forms of anxiety related to ‘Moroccan’ Dutch were mentioned to explain why ‘Moroccan’ Dutch have a disadvantaged labour market position.
17
Anxiety about criminality
‘Moroccan’ Dutch men are often associated with criminal behaviour. Fear that an employee will steal, behave aggressively or show deviant behaviour is mostly mentioned as a reason for the average employers to not invite the ‘Moroccan’ Dutch candidate. Most respondents share this opinion to some extent. The rushed to clarify they do not want to generalize and stress that these expectations are certainly not applicable to all ‘Moroccan’ Dutch men. Nevertheless, they underline the negative reputation of ‘Moroccan’ Dutch men and most respondents endorse the idea that the risk of criminal behaviour is higher when hiring a ‘Moroccan’ Dutch compared to an ‘autochthonous’ Dutch. The association of ‘Moroccan’ Dutch with criminality and unreliability was mentioned by employers in 2010 as well (Nievers & Andriessen, 2010, p. 17).
None of the respondents had experienced criminal behaviour of ‘Moroccan’ Dutch men in their own business. The stereotypes are sometimes based upon experiences outside work, mainly on media reports or statistics about criminality. ‘Moroccan’ Dutch are overrepresented in police reports as suspects in comparison to ‘autochthonous’ Dutch (CBS, 2014, p. 71) and this has gained a lot attention in media. Often employers referred to these criminality statistics or media coverage:
The group has a negative image. I don’t want to generalize too much, but look at Opsporing Verzocht [a program that shows not yet arrested suspects], all those television programmes. There is almost no white person mentioned. Off course, that is not the whole story. But it gives an idea. An idea of thought. [What is that idea?] Bag robbers, boys who beat other boys up, or who beat girls. Look at all those clips or television. Youtube, those kind of things. Yes, it will not be always Moroccans, but the whole group has a negative image. (hospitality, R2)
This citation clearly illustrates the mechanism of statistical discrimination. Characteristics of a group – in this case overrepresentation in police statistics – are ascribed to individual group members. The productivity or appropriateness of a candidate is estimated with this information. Employers want to avoid risks and strive mainly for maximizing their profit. When hiring a ‘Moroccan’ Dutch candidate instead of an ‘autochthonous’ Dutch, the risk of hiring a less productive or less appropriate candidate is perceived to be higher. Although employers know rationally that not all ‘Moroccan’ Dutch are criminals or cause problems, they perceive the risk of experiencing difficulties as higher, within this group. Following Elias and Scotson (1965), the characteristics of the worst members of the ‘Moroccan’ Dutch group – individuals who are involved in criminality – are generalized to the whole group.
18
Anxiety about customers’ opinion: headscarf
A Moroccan female name evoked associations with the Islam and the likelihood of her wearing a headscarf. The majority of the employers felt more or less uncomfortable with inviting a ‘Moroccan’ Dutch headscarf wearing woman and expected the average employer to feel the same. Most of the respondents indicated never having received a job application from a woman with a headscarf. Nevertheless, they mentioned it as something they would carefully think about. No statistical discrimination arguments were given about the expected productivity of these women, but taste-based reasons (Becker, 1957) – for themselves and for customers – were reported.
The main reason why employers felt uncomfortable in hiring ‘Moroccan’ Dutch women with a headscarf is that they are afraid this personnel would influence their reputation and the customers they address. They perceive a headscarf as not compatible with how they want to present their business; a headscarf does not fit the image they want to convey. In a clothing shop with a young and hip audience, the employer explained that a headscarf is not in line with the spontaneous reputation of the shop. Furthermore, the shop does not aim to address Islamic customers.
Well, yes, we might invite them, but actually it is not what we want to convey. No, it is not. It is because…. Well, actually, they are not the customers we would like to have. So maybe, if someone with a headscarf is working here, you also get all those customers. And I think they [higher management] wants to prevent this. (retail, R7).
For me, a headscarf is the same as a tattoo, you have to cover it. I prefer that you don’t have it. Same for large necklaces. […] I have the opinion that waiting staff should always be present at the background. With such externals, you attract attention. And actually, I don’t want that. I want that we are always available, but at the same time at the background. We have to be neutral […] It is like background music for example. And yes, someone with a headscarf draws attention. And people would have an opinion about it and I find that annoying, I think. (hospitality, R6)
The last quote mentioned the image an employer wants to convey with their business as well. The employers want to create a specific atmosphere and a woman with a headscarf would not fit into this picture. Interestingly, the last quote adds the expected taste of customers to her reasoning. The possibility that her customers “would have an opinion” about headscarves, whereas she want to be her personnel neutral, is perceived as problematic for this respondent. Saharso (2007, p. 519) shows that the public debate around headscarves in the Netherlands is centred around the concept of neutrality and also the following quote has non-neutrality of the headscarf as topic. Here, a headscarf is perceived as not matching with corporate clothing and is framed as deviating:
19 No, I would not accept it [headscarf]. [Why?] Just appearance. How should I imagine that? I just cannot imagine it, someone with corporate clothing and a headscarf… But last time, we had someone who painted her hair red. That is also not accepted. I think that is actually exactly the same […] I think it is also related to norms and values in a company, and rules. [Are there rules about religious externals?] No, it is not in our house rules. (hospitality, R9)
A symbolic or categorical boundary is drawn between women with and women without a headscarf. The former is seen as another – non-neutral – social category. Customers are expected to prefer that companies do not hire women with a headscarf. This taste of customers is often constructed by the employers themselves. Only one respondent indicated that in his supermarket customers sometimes explicitly complain if all cashiers were headscarf wearing women. In other cases, respondents constructed the opinion of their customers themselves. Nievers and Andriessen (2010, p. 16) have similarly concluded that especially in functions with customer contact, employers view woman with a headscarf as problematic.
However, the expected taste of customers is not the only form of taste-based arguments that were given. The following quotes show that the taste of the employer itself is important as well. The first quote is from the employer who most strongly argued in favour of ‘autochthonous’ Dutch candidates. Although the respondent starts her argument with indicating that she would lose her customers, it turns out to be her own taste that is crucial. In the second quotation it is clear that the employers’ own taste is the issue. The third quote illustrates that the taste or opinions of customers was mentioned to legitimize the unease of the employer. However, the reasoning was not fully developed and here it seems that relying on the taste of customers is also a comfortable way to follow one’s own preferences.
[Would you hire a woman with a headscarf to work here as a waitress?] No, absolutely not! That would be horrible! Then I will present myself in such way, no, my cafe would be empty if I had a girl with a headscarf. [Because the customers don’t want it?] Well, I don’t want it! No, absolutely not. I cannot imagine that they will work here with headscarves. [And why is that so bad?]. Well, that’s their decision, but not here. I don’t want to be a sort of Moroccan coffeehouse! (hospitality, R12)
I have no headscarf wearing women. I think it does not fit my shop. I am not racist, absolutely not. I just had this idea when once, a couple of years ago, so many women with headscarves entered my shop and I felt very intimidated. […] It is just a feeling. (retail, R17)
Bouchra is probably someone who wears a headscarf. And yes, my opinion is that in my company… Well, there are many companies, like supermarkets, all the cashiers wear headscarves. Absolutely no problem. But my guests are in general all white Dutch people.
20 And I don’t think that they have a problem with it, but for me… In my atmosphere… Well, actually, I never been in such a situation. (hospitality, R2)
Another last important aspect is that women with a headscarf are expected to not fit into the team. We saw earlier that this is a highly important criterion. Employers remain however vague about why women with headscarves would fit in less. Again, it is more based upon their feelings and they can hardly explain it. It seems that women wearing headscarves are perceived as culturally very different by most employers, although they could not pinpoint what these differences consist of.
Anxiety about cultural differences
In addition to anxiety for criminal men and women wearing headscarves, employers mentioned a fear for cultural differences. ‘Moroccan’ Dutch candidates are seen as the ‘cultural other’ in comparison to ‘autochthonous’ Dutch. Culture is used as an umbrella term and different aspects were brought up. Firstly, employers emphasize the importance of language proficiency. Several employers stated that they considered language proficiency important for functions with customer contact, but less for functions without, e.g. cooks. Demanding a good knowledge of the Dutch language is not discrimination per se. If good language proficiency is necessary to perform the job well, an employer may require it. It is problematic and discriminatory, however, if employers already expect that language proficiency is not sufficient when a ‘Moroccan’ Dutch candidate applies. In that case, again, potential group characteristics are ascribed to an individual member.
Related to language proficiency is the fear of that communication problems could occur between personnel. These problems are caused by differences in interpretation. Employers who approach their employees in a direct manner can sometimes be misunderstood. The expectation is that jokes about family escalate more quickly with ‘Moroccan’ Dutch employees than with ‘autochthonous’ Dutch.
Thirdly, several employers denominate a cultural difference in interaction between men and women. They stated that ‘Moroccan’ Dutch men can find it difficult to have a female manager or work pleasantly together with female colleagues, because they do not treat them with respect. An employer explains:
I work a lot with women, so this [Moroccan name] should be someone who is a bit progressive in comparison to his culture. Because otherwise, I could maybe expect problems. […] Most Moroccans interact differently with women than what we consider, on average, normal. They rather listen to a man and that’s a problem here. So yes, they should be able to do so. And that isn’t something they have learned at home. (hospitality, R3)
21 Lastly, some employers mentioned a cultural difference in work attitude. ‘Moroccan’ Dutch are by some perceived as stubborn or people who try to delegate their work to others. Another respondent argued that he experienced that ‘Moroccan’ Dutch employees are often too late and not disciplined enough:
I think it’s still a certain culture they inhabit. And well, for a Dutch person and even for a Surinamese, it is much easier for them to make that switch to what we expect, than for a Moroccan. [What is the difference in culture?] Well, my experience with Moroccan boys; they are too nonchalant and discipline is not sufficiently available. (retail, R20)
Cultural differences are thus mentioned as a reason to prefer not to hire ‘Moroccan’ Dutch. Duyvendak et al. (2009) show that culture is highly important in the Dutch integration debate. Especially Islamic minority groups are more and more perceived as culturally different. Majority populations increasingly fear Muslims. Duyvendak posed that we have seen a culturalization of citizenship, that is “a process in which emotions, feelings, norms and values, symbols and traditions (including religion) come to play a pivotal role in defining what can be expected of a Dutch citizen” (Duyvendak, 2011, p. 92). He showed that it is expected that ‘migrant’ groups adjust to Dutch culture and participate. Siebers and Dennissen (2015) argue similarly that the Dutch migrant discourse can be described as cultural fundamentalism wherein the culture of ‘autochthonous’ Dutch and primarily Islamic ‘migrant’ groups are perceived as incompatible. The above quotes show this focus on the cultural differences and the perceived incompatibility.
Anxiety caused by Islamic State
A last form of anxiety related to ‘Moroccan’ Dutch candidates is influenced by developments in the Middle East in general and Syria in particular. A small number of employers referred to the rise of Islamic State and explained how this negatively influenced the image of ‘Moroccan’ Dutch in the Netherlands.
Well, I think that for the Moroccan people it is more… Also because all those situations right now! Because of Islamic State. People are afraid because of that. Really afraid. For Hindustani people that fear does not exist. I think yes, that the fear related to a Moroccan name is bigger. (hospitality, R13)
Yes, fear for foreigners… It is just, look at what happened in the world. If you see that Islamic State has taken and converted half of Syria yesterday, people link that to the Islam. Unfortunately, the extremist Islam. Yes, that is a good example of what happens now. And unfortunately the extremists mess up the image of the rest of the Muslims. (hospitality, R11)
