Determining discrimination. A multi-method study of employment discrimination among descedants of immigrants in Norway.

After decades of research documenting the various processes of marginalisation and exclusion immigrants experience in their European countries of residence, recent efforts have been made to explore the extent to which these disadvantages are transferred across generations. Illuminating the barriers facing descendants of immigrants when trying to access labour market opportunities is a task of growing importance, and the ‘second generation’ – children of immigrants either born in their parents’ destination country or immigrating before adolescence – makes up a particularly relevant case for discrimination research. First, this group of individuals usually speaks the majority language fluently, and they have acquired domestic educational merits and work experience. As such, important obstacles normally assumed to explain many of the disadvantages facing immigrants do not apply for the second generation, making discrimination in hiring easier to detect. Second, although there are differences between groups, children of immigrants have on average achieved impressive results in the educational systems across Europe. If they are not offered the opportunity to translate their educational investments into relevant work, this has consequences both at the individual level and for the society as a whole. For individuals, barriers to employment may lead to social and economic marginalisation. At the societal level, widespread discrimination is economically inefficient and potentially a source of social unrest. Hence, the relevance of studying patterns of discrimination among descendants of immigrants is beyond doubt. Although it is important to assess the extent to which children of immigrants experience discrimination in employment, the research on second-generation incorporation – like the field of discrimination research itself – suffers from methodological problems in measuring the extent and causes of disadvantage. In traditional statistical approaches, discrimination can hardly be distinguished from social network effects; in employer surveys, the relationship between accounts and practices remain uncertain; in laboratory experiments, findings are not directly transferable to empirical realities; and in ethnographic field work among potential victims of discrimination, the representativeness of the exclusionary processes reported is notoriously unclear. Furthermore, theoretical models used in economics and quantitatively-oriented sociology have a tendency to favour single-factor explanations of discrimination at the individual level, not sufficiently addressing how organisational contexts may shape patterns of exclusion above and beyond individual motives and cognitive biases. This dissertation addresses some of these problems in the discrimination literature by complementing a field experiment on discrimination in the Norwegian labour market with indepth employer interviews. In the first part of the study, a large-scale field experiment, in which hundreds of fictitious, paired résumés and cover letters were sent in response to real job openings, was used to measure the extent to which children of Pakistani immigrants are discriminated at the entrance to the labour market in the greater Oslo area. Because the two fictitious job candidates in each pair were equally qualified in every productivity-relevant aspect, but were randomly assigned a Norwegian or a Pakistani name, the direct effect of ethnic background on job interview offers is isolated. Hence, the experimental approach allows for a direct measure of ethnic discrimination in hiring processes. The field experiment leaves little doubt that descendants of immigrants indeed suffer from discrimination in access to employment in Norway: For the study in total, the probability of receiving a job interview offer is reduced by 25 per cent for the minority applicant compared to the equally qualified majority applicant. However, there are important differences within these overall results. For example, the discrimination rates are larger in the private sector than in the public sector, and there are significant differences across the occupations included in the study. These variations indicate the occurrence of different processes of exclusion at different locations in the labour market, pointing to the need for context-sensitive interpretations of the results from field experiments. In the second part of the study, in-depth interviews with a subsample of the employers participating in the experiment explore how, when, and why the ethnic background of job applicants comes to matter in decision-making processes. Supplementing the field experiment with employer interviews rests on a theoretical assumption that the field experiment literature has been too concerned with single-factor explanations at the individual level, in which discrimination either is caused by consciously acting employers or is due to cognitive bias. Conducting interviews with employers that received the fictitious résumés in the first stage of the study allows for a qualitative exploration of the hiring processes, acknowledging that although the experiment suggests a causal relationship between ethnic background and employment opportunities, there are several ways in which a discriminatory outcome may be produced. Indeed, the interviews suggest the need for multi-level explanations. At the individual level, many employers use fixed images of the ‘immigrant’ when assessing the quality of applicants with foreign names, regardless of whether the applicants are of the first or second generation. As economic models of statistical discrimination assume that employers use accurate depictions of the average productivity level among different groups when considering job applicants – and there indeed are large differences in group productivity between the generations – the tendency to equate a foreign name with stereotypes attached to the immigrant experience supports social-psychological research on stereotypes and biases in recruitment. However, the interviews also point to the relevance of explanations at the organisational level. The qualitative ‘tracing’ of recruitment processes reveals what seems to be an interaction between the context of employment and the outcome of hiring decisions, indicating that means of bureaucratisation (e.g. formalised recruitment procedures limiting the room for employers’ discretion) may serve as a lever against discrimination in hiring. These results are in line with insights from organisational-level theories of workplace inequality, and they illustrate the relevance of combining field experiments with qualitative methods to better grasp the factors shaping labour market opportunities in modern societies. The dissertation consists of two main parts. The first part is an introductory chapter, presenting the main objectives of the research, as well as reflecting on the theoretical, methodological, and ethical underpinnings of the study. The second part consists of four scholarly articles. The first article reviews important methodological debates within the field experiment literature and presents the particular research design of this study. The next three articles discuss the main empirical findings and their theoretical implications