generalization or aversion? – UGent @ Work – Ghent University

PhD student Louis Lippens

(June 10, 2022) After the fourth and fifth Story @ UGent @ Work, there was a lot to do about inactivity among 25-64 year olds in Belgium with non-EU27 nationality. 44.2% of this group were neither in employment nor job seekers. It places our country at the very back of the European rankings. A problematic fact in light of the Flemish, Walloon and federal government’s intention to get more people into work. In this blog post, we discuss one of the possible explanations for this high inactivity on the employer side: recruitment discrimination. More specifically, the scientific evidence for the mechanisms of recruitment discrimination is discussed, and some solutions are presented to combat discrimination.

Recruitment discrimination, unequal and disadvantageous treatment of individuals based on attributable characteristics (such as national origin), is a well-documented phenomenon. In academic research, we typically measure recruitment discrimination by comparing the positive responses to applications from similar candidates, which typically differ in only one discrimination characteristic. Based on worldwide research, we know that people of a different ethnicity or origin on average receive a third less positive response. Depending on the specific origin or ethnicity, this difference even averages almost 50%. The figures for age discrimination are the same: Older candidates receive on average about 40% less positive response than equal, younger candidates.

The reduced employment opportunities again entail economic and social costs. On the one hand, there is an increased risk of long-term unemployment and inactivity for graduate employees from the various minority groups. This leads to underutilization of talent. On the other hand, organizations may not be able to attract the most suitable candidates or recruit inadequate candidates due to discrimination in the recruitment phase. Especially in times of labor shortage, discriminatory employers may have extra difficulty in attracting appropriate talent, especially when discriminating on the basis of personal preferences. In extreme cases, this can even lead to (accelerating) corporate tackling, as has already been confirmed in US research.

Selection based on statistics

A first discrimination mechanism, statistical discrimination, is based on employers making election decisions based on group information. An employer who, based on previous collaborations with staff from an ethnic minority group, believes that people from this group have a lower command of Dutch, can use this information to exclude an individual candidate with the same ethnic characteristics from the selection process at an early stage. The reason for the exclusion is then based on the fact that it is statistically probable that this candidate also masters Dutch poorly. Only in case the candidate does not actually speak the language and the language is a relevant selection criterion, can it be a correct decision; otherwise it is almost always discrimination. Furthermore, without subjecting the candidate to a language assessment, it is impossible to determine whether the decision was justified.

This mechanism is not limited to recruitment discrimination based on ethnicity. We also see a similar accountability mechanism when it comes to age discrimination. For example, there are a number of beliefs about older workers that limit their opportunities in the labor market: They are seen as less technologically skilled, less flexible and less educated. In addition, the perceived high labor costs associated with recruiting older workers are also an obstacle for some employers.

However, an important caveat must be made with regard to the above-mentioned discrimination mechanism. It is not necessarily the case that the beliefs one has about a particular group in the labor market are correct. Also in economic-scientific research, it is often assumed that the information that employers have about, for example, employees of different backgrounds or older employees, and which they use in the selection process, is correct. However, candidates may be treated unequally and negatively on the basis of incorrect beliefs about group characteristics.

Dislikes in the hiring process

An alternative discrimination mechanism that can also provide an explanation for the perceived labor market discrimination is the preferential discrimination mechanism. Preference discrimination is characterized by a reluctance to work with people from a particular minority group. This reluctance may be motivated by the employer himself, but also by employees or customers who do not want to work or interact with people from a particular minority group. The employer’s (financial) sensitivity to the wishes of employees and customers can subsequently lead to reduced opportunities in the labor market.

Based on research, we see that both statistical discrimination and preferential discrimination play a role in the field of ethnic labor market discrimination. Although the mechanism of statistical discrimination is sometimes invoked to justify discrimination, it is certainly not always this mechanism that drives ethnic discrimination. In fact, there is scientific evidence that ethnic discrimination in the employment phase is primarily motivated by employers’ personal preferences or fears that their employees or customers prefer not to work or interact with people of different ethnicity or national origin.

Solutions to combat discrimination

From an economic perspective, there are a number of solutions to combat discrimination in the labor market. A logical counter-reaction to reducing statistical discrimination is to obtain more (accurate) information about the individual candidate’s potential. This should prevent the return to group characteristics or statistics to estimate the productivity of graduate staff. On the one hand, it is the candidate’s responsibility to provide adequate information about himself in the initial phase of the selection process so that the employer can make an informed decision. On the other hand, it is in the employer’s interest to obtain this information in order to select the most suitable candidate for the job. This can be done by structuring the selection process more or by training employers or recruiters to suppress stereotypes and focus on the candidate characteristics that are relevant to the job they are recruiting for.

An appropriate measure to combat preferential discrimination is to increase the cost of discrimination. Discrimination based solely on preference or aversion is a rather economically irrational form of discrimination. Increasing the cost of discrimination should therefore neutralize the perceived burden of collaborating or interacting with minority people. Based on research, we see that punishing such discrimination is also a little more effective than rewarding non-discrimination. The fact that the mechanism of preferential discrimination is likely to play a role in recruitment decisions suggests that in clear cases of discrimination for which there is sufficient evidence, effective sanctions should also be imposed to reduce discrimination in the recruitment process. In practice, this means, firstly, that existing anti-discrimination legislation should be better applied.

By PhD student Louis Lippens (Department of Economics; Louis.Lippens@UGent.be

Read more about this research in the published scientific article

Also read the other UGent @ Work blog posts.

Leave a Comment