Periodic Reporting for period 4 - DIVERSE-EXPECON (Discriminative preferences and fairness ideals in diverse societies: An ‘experimental economics’ approach)
Reporting period: 2022-07-01 to 2023-12-31
Ethnic minorities are not always treated favorably. Several factors potentially underlie this achievement gap and ethnic discrimination is one of these factors. To illustrate, job applicants in the EU with a non-Western background have to send more than twice as many applications to get invited to a job interview than persons belonging to the majority group, who have an equivalent curriculum vitae. In general, there are two forms of discrimination: statistical and taste-based. Statistical discrimination refers to discrimination in a context with strategic uncertainty. For example, an employer who is uncertain about the future productivity of a person with a different ethnicity forms beliefs relying on general information about the different ethnic group to which this person belongs. If he believes that certain ethnic minorities are less productive, he may be reluctant to recruit from this group. Taste-based discrimination refers to discrimination in a context without strategic uncertainty. Suppose that the employer would be perfectly aware that the future productivity of all job applicants is exactly the same and is still less willing to hire a person with an ethnically different background, then we say that this employer discriminates on the basis of tastes. This project studies taste-based discrimination in a set of European countries. For the design of policies that maximize total welfare or reduce discrimination it is important to understand the mechanisms behind discrimination.
The project consisted of three main objectives: (i) Investigating whether and how preferences of European natives in social interactions depend on others’ ethnicity; (ii) Studying whether natives have different fairness ideals when it comes to natives than when it comes to non-natives; (iii) Analysing whether ingroup preferences and fairness ideals depend on exposure to diversity.
The project consisted of three main objectives: (i) Investigating whether and how preferences of European natives in social interactions depend on others’ ethnicity; (ii) Studying whether natives have different fairness ideals when it comes to natives than when it comes to non-natives; (iii) Analysing whether ingroup preferences and fairness ideals depend on exposure to diversity.
To investigate whether and how preferences of European natives in social interactions depend on others’ ethnicity, experiments were conducted in the Netherlands, France and Germany in which participants belonging to the majority population (the decision makers) made choices that affected another participant. This other participant either had a majority background (like the decision makers) or belonged to an ethnic minority group. Whether or not ethnic discrimination took place was investigated by comparing choices of decision makers matched to another majority person to those of decision makers matched to a person with a minority background. In an experiment conducted in the Netherlands it was found that native Dutch decision makers were less likely to reciprocate trust by a trustor with an ethnic minority background than by a trustor with a majority background, leading minority trustors to achieve a lower return on trust than majority trustors (see Figure 1). Experiments in the Netherlands, France and Germany in which decision makers made simple allocation choices in which they chose how to divide money among themselves and another (passive) participant did, in contrast, not lead to substantial discrimination. By connecting the data from different experiments and from different waves of the same experiments to ethnicity-related events such as the European refugee crisis and the #BlackLivesMatter protests, we found these events had marked influence on choices made by decision makers in the experiments, but only on those decision makers whose choice affected someone with an ethnic minority background. One paper was published in the Economic Journal and one paper as a CEPR working paper; two more working papers are currently being written to be submitted to academic journals.
To study whether natives have different fairness ideals when it comes to other natives than when it comes to ethnic minorities, general-population experiments were carried out in which participants (the decision makers) decided about the distribution of money between two "workers" who had carried out a productive task. One of these workers had a majority background (the same as the decision maker) and the other worker belonged to an ethnic minority group. The experiments were conducted in France, Germany and Belgium. In part of the experiments both workers were equally productive, whereas in other experiments there was a productivity difference between the two. A general finding from these experiments is that the decision makers employed a double fairness standard; on average, more money was redistributed away from a minority worker to a majority worker than vice versa. One paper (a literature review) was published in the Journal of Economic Surveys and two working papers are currently being written to be submitted to academic journals.
Finally, to investigate whether ingroup preferences depend on exposure to diversity, the team employed two different methods that varied in the extent of naturalness. The first method was that of an impersonal and anonymous laboratory experiment carried out with student participants with the aim of studying the effect of cooperative versus competitive interactions with in- or out-group others on discrimination. The research found that cooperative interaction between individuals, even though impersonal and anonymous, eliminated any in-group bias as compared to the case where there was no such interaction; a competitive interaction reduced pro-sociality towards others in general, irrespective of whether these others were in- or out-group. The second method was a state-of-the-art econometric analysis of individual-level data assembling political preferences and attitudes to ethnic diversity and geographical data on asylum seeker centers from the COA (Central Organ for Asylum seekers in the NL). The aim was to study the causal effect of exposure to ethnic minorities on political preferences and attitudes towards them, finding that individuals who got exposed to asylum seekers in their close neighborhood during the so-called European refugee crisis in 2015-2016 developed a more positive attitudes to ethnic diversity and were less inclined to vote for far-right political parties than individuals in neighborhoods without asylum seekers. One paper was published in the Journal of Economic Behavior & Organizations, another paper is conditionally accepted for the Economic Journal, and a third (spin-off) paper is currently in preparation.
The researchers on the team have presented the project's results in numerous invited seminars at universities across Europe, at leading international academic conferences across the world, with as a highlight a keynote speech at the annual world conference of the Economic Science Association in 2019, and at conferences attended by policy makers, with as a highlight a keynote speech at the Royal Dutch Economic Association New Paper Sessions in 2021. The research founds its way to open-access policy journals (e.g. EU Research) and was referenced in high-quality media outlets in the Netherlands (De Correspondent and NRC) as evidence showing that the new Dutch legislation regarding the spreading of asylum seekers (the "Spreidingswet", in the mean time approved by the Parliament) should not raise concerns.
To study whether natives have different fairness ideals when it comes to other natives than when it comes to ethnic minorities, general-population experiments were carried out in which participants (the decision makers) decided about the distribution of money between two "workers" who had carried out a productive task. One of these workers had a majority background (the same as the decision maker) and the other worker belonged to an ethnic minority group. The experiments were conducted in France, Germany and Belgium. In part of the experiments both workers were equally productive, whereas in other experiments there was a productivity difference between the two. A general finding from these experiments is that the decision makers employed a double fairness standard; on average, more money was redistributed away from a minority worker to a majority worker than vice versa. One paper (a literature review) was published in the Journal of Economic Surveys and two working papers are currently being written to be submitted to academic journals.
Finally, to investigate whether ingroup preferences depend on exposure to diversity, the team employed two different methods that varied in the extent of naturalness. The first method was that of an impersonal and anonymous laboratory experiment carried out with student participants with the aim of studying the effect of cooperative versus competitive interactions with in- or out-group others on discrimination. The research found that cooperative interaction between individuals, even though impersonal and anonymous, eliminated any in-group bias as compared to the case where there was no such interaction; a competitive interaction reduced pro-sociality towards others in general, irrespective of whether these others were in- or out-group. The second method was a state-of-the-art econometric analysis of individual-level data assembling political preferences and attitudes to ethnic diversity and geographical data on asylum seeker centers from the COA (Central Organ for Asylum seekers in the NL). The aim was to study the causal effect of exposure to ethnic minorities on political preferences and attitudes towards them, finding that individuals who got exposed to asylum seekers in their close neighborhood during the so-called European refugee crisis in 2015-2016 developed a more positive attitudes to ethnic diversity and were less inclined to vote for far-right political parties than individuals in neighborhoods without asylum seekers. One paper was published in the Journal of Economic Behavior & Organizations, another paper is conditionally accepted for the Economic Journal, and a third (spin-off) paper is currently in preparation.
The researchers on the team have presented the project's results in numerous invited seminars at universities across Europe, at leading international academic conferences across the world, with as a highlight a keynote speech at the annual world conference of the Economic Science Association in 2019, and at conferences attended by policy makers, with as a highlight a keynote speech at the Royal Dutch Economic Association New Paper Sessions in 2021. The research founds its way to open-access policy journals (e.g. EU Research) and was referenced in high-quality media outlets in the Netherlands (De Correspondent and NRC) as evidence showing that the new Dutch legislation regarding the spreading of asylum seekers (the "Spreidingswet", in the mean time approved by the Parliament) should not raise concerns.
The research constitutes the first systematic data collection on taste-based discrimination in Europe (see previous section).