Over at the NCCR on the move blog, there’s a summary of research undertaken jointly with Eva Van Belle on CV Whitening. The blog post takes a step back and embeds considerations of CV Whitening in broader research on ethnic discrimination in the job market and the correspondence tests we use to measure discrimination.
In Western countries, we observe that some employers discriminate against people with a name indicating that they are from a minority group. Research has shown that if you have a name signalling that you are an ‘immigrant’ or belong to an ‘ethnic minority,’ the chances of being invited to a job interview are reduced. Factoring in the potential discrimination, some immigrants and members of ethnic minority groups have been observed to adopt strategies to hide details that show their minority status from their applications.
Correspondence tests have highlighted systematic discrimination in the labour market. In a correspondence test researchers create fictitious applications and send them to real employers and landlords. The applications are constructed in a way to highlight discriminatory practices in cases where minority applicants are invited less often to interviews. On average, applicants from non-neighbouring countries find it much harder to be invited to a job interview.
Considering the discrimination, ‘immigrants’ and members of ethnic minority groups have a clear incentive to hide details from their applications that make their minority statuses apparent. On the contrary, if an employer cannot tell that the applicant is an ‘immigrant’ or belongs to a minority group, the chances of getting invited to an interview have been noted to increase.
Our publication on hiring discrimination against Blacks in Switzerland is now properly published at JEMS. Using a correspondence test, we find the unfortunately usual pattern of discrimination in Switzerland, too.
I’m very happy to announce a new publication in JEMS on hiring discrimination of Blacks in Switzerland (joint work with Rosita Fibbi, Eva Zschirnt, and Robin Stünzi). Sometimes it’s funny how events unfold — the decision to run this correspondence test to measure hiring discrimination on the basis of skin colour was taken in 2014, and we went into the field in 2018. Then, in 2020, the Black Lives Matter movement gained momentum and increasingly people in Western Europe started talking about racism and discrimination against Blacks. In this sense, we‘re very happy to make our contribution to document discrimination as an undeniable fact, regardless of the fact that some keep claiming that without formal colonies Switzerland could not be affected by racism (I’ll leave the “happy” for times when things get better).
We show that Black jobseekers in Switzerland must send around 30 per cent more applications than White candidates in order to be invited to a job interview.
Not entirely by coincidence, we can compare the results with other recent correspondence studies in Switzerland that cover immigrants from Kosovo, and we can show that the level of discrimination is substantively equivalent for applicants with a Kosovo-Albanian name. This suggests that in the Swiss case there is on average no additional penalty for skin colour. Explorations, however, reveal significant differences in discrimination rates between urban and rural settings, opening new avenues for understanding why ethnic and racial discrimination vary across geographical contexts.
Rosita Fibbi, Didier Ruedin, Robin Stünzi & Eva Zschirnt (2021) Hiring discrimination on the basis of skin colour? A correspondence test in Switzerland, Journal of Ethnic and Migration Studies, DOI: 10.1080/1369183X.2021.1999795
Here’s an exciting new study on hiring discrimination. They got access to the behavioural data of online recruiters to find evidence of discrimination against atypical candidates: Contact rates by recruiters are 4–19% lower for individuals from immigrant and minority ethnic groups, depending on their country of origin, than for citizens from the majority group. Women experience a penalty of 7% in professions that are dominated by men, and the opposite pattern emerges for men in professions that are dominated by women.
I find it interesting that they pitch their method as an alternative to correspondence tests (perhaps not all that novel if we’re looking outside the strict focus on hiring discrimination). We’re seeing an increasing number of correspondence tests in recent years, despite important ethical concerns. Not all of them are reasonably motivated, in my view — “no recent correspondence test” in a particular country/for a particular group/occupation does not cut it for me –, but jointly these studies give us a pretty clear picture of discrimination (especially in Western countries). Access to recruiting databases may not be possible in all countries, and we’re still struggling with the blatant omission of informal labour markets and internal recruitment. On the other hand, at least in principle we could test different interfaces and see if we can reduce discrimination this way…
A new correspondence study from Denmark shows discrimination by school administrators against parents with ‘Muslim’ names. They sent letters to schools across the country to ask whether they could move their son to that particular school (implying that they were not happy with the current school). 25% of fathers with a ‘Danish’ name (i.e. Peter Nielsen) received a positive answer, compared with 15% of fathers with a ‘Muslim’ name (i.e. Mohammad Osman).
In addition to holding everything constant by using men only (fathers enquiring about their sons), they had a variation in whether the son was a ‘diligent’ student. An interesting qualitative detail is that ‘Muslims’ are more often subjected to additional questions by e-mail (simple questions like verifying they actually live in the catchment area), while the ‘Danes’ were more often asked to call.
I find it interesting that their point of reference were studies on discrimination by public officials (typically politicians), but did not reflect methodological innovations from other correspondence tests, like stimulus sampling (!), or considerations of unmatched designs. I find it disappointing to find that the pre-registration at EGAP leads to a “page not found” error, especially since footnote 1 contains this interesting teaser: “We diverge from the preregistration to limit our focus only to the two variables that were subject to experimental manipulation and causal inference rather than those conditional on posttreatment responses.”