Although discrimination continues to be ubiquitous in labour markets, it is very difficult to measure the true extent of discrimination because, for obvious reasons, respondents are deliberately or subconsciously inclined to conceal it. Thus, it was not until two or three decades ago that researchers managed to develop and implement a robust methodology for studying discrimination in labour markets. This methodology, known as a resume audit study, involves sending out pairs of almost identical fictitious resumes, typically with one crucial difference that indicates a particular racial/gender/ethnic group, and measuring the likelihood of receiving requests for an interview.

In one of the most famous audit studies conducted in the US1 , economists sent out resumes with names that were typical of different ethnic origins and found that those with Asian and African sounding names, such as Lakisha and Jamal, for example, are 50% less likely to receive an interview invitation than those with white sounding names, such as Emily and Greg. Surprisingly, the researchers found that discrimination exists even in organisations that declare themselves to be equal opportunity employers.

In another study, Israeli researchers tested the effect of facial beauty by adding photographs to CVs. Their most surprising finding was the penalisation of attractive women compared to women who do not include a headshot with their CV.

The following two possible explanations merit attention, and are based on the beauty premium hypothesis2, according to which more attractive women are more likely to be promoted and earn a higher income. One is based on the so-called “dumb-blonde” stereotype, which pervades Western culture, which suggests that attractive women, represented here by blondes, are able to rely on their looks to advance their careers and thus do not make use of their intelligence. Knowing this, employers would be reluctant to invite attractive women for an interview. The second possible explanation is that CVs are often initially screened by HR officers, the majority of whom are traditionally women. These women may view other attractive women as potential competition or threat, creating a higher entry barrier for attractive female applicants.

One policy implication stemming from these audit studies is making job applications anonymous, with names, gender, and photographs removed, so that the individual reviewing and shortlisting candidates does not have any information that may prompt discrimination. However, a very recent study3, using data from France, reported that even this anonymization does not help resolve this issue. In fact, with “blind” hiring procedures, the interview invitation rate for minority candidates decreases, while that of majority candidates increases. The researchers argue that the participating firms tended to interview and hire relatively more minority candidates to begin with, i.e. prior to the implementation of anonymization. Thus, anonymization prevented them from treating minority candidates more favourably during the blind process.

So how can companies ensure that discrimination ceases to exist? For now, this question remains unanswered.