An official website of the United States government
Beyond BLS briefly summarizes articles, reports, working papers, and other works published outside BLS on broad topics of interest to MLR readers.
The impact of workplace discrimination has been studied for at least the past 60 years. Much of the earliest research assumed that employers’ biases did not directly affect output. But a new line of research examining the adverse impact of stereotypes about minorities on worker performance is tearing down this assumption. In a recent study of this phenomenon, Discrimination as a self-fulfilling prophecy: evidence from French grocery stores (National Bureau of Economic Research, working paper 22786, October 2016), Dylan Glover, Amanda Pallais, and William Pariente find that managers’ bias does indeed affect the performance of minority workers.
The researchers use the Implicit Association Test (IAT) score to measure managers’ bias toward ethnic minority workers. The IAT is a social psychology test designed to measure an individual’s unconscious biases toward different demographic groups. The IAT can be tailored to specific characteristics, such as race, gender, and ethnicity. For this study, the IAT measured managers’ bias toward North African-sounding names. (Because French law prohibits asking workers their ethnicity, workers’ names were used as an indicator of minority status in the study.) Managers who scored poorest were more likely to associate such names with worker incompetence.
Research was conducted in 34 outlets of a French grocery store chain with a large proportion of minority workers. For each of the stores, the researchers tracked the work performance—absences, time worked, scanning speed, and time between customers—of all new cashiers (those working under the initial 6-month contract). Using both the IAT data and the worker performance data, they found that minority workers performed worse on days they were scheduled to work under biased managers.
When minority workers were scheduled to work shifts with biased managers, workers’ performance suffered in a variety of ways: they were more likely to be absent, they spent less time in the store (generally because they were less likely to stay after their shift ended), and they performed their work at a slower rate. This effect was more pronounced in stores with fewer minority workers. Because these are hourly employees paid on the basis of time worked, the researchers suggest that one of the effects of working for biased managers is “a loss in wages for minority workers.”
One of the most encouraging findings was that biased managers did not display any animus toward minority workers. That is, they did not treat minority workers worse than other workers. In fact, biased managers “were less likely to assign [minority workers] unpleasant tasks (cleaning), and no more likely to assign them unpleasant registers or breaks.” And the workers themselves did not report disliking or feeling disliked by the biased managers. Given this, the researchers suggest that the negative effects of the bias may be the result of implicit bias constraining worker–manager interactions. That is, biased managers choose to interact less with minority workers and, accordingly, ask less of those workers.
The researchers acknowledge a clear limitation of the study, which is that the results relate to a single type of worker in a single, specific setting—cashiers in French grocery stores. It is not clear how applicable these findings are to other types of workers, in other workplace environments, and in other nations. Another limitation worth noting involves the controversial IAT. There is a disagreement among some researchers about whether the IAT helps predict the likelihood of someone showing implicit bias against members of marginalized groups.
Under the assumption that the results are reflective of manager bias on performance, the researchers suggest that workplaces adopt specific policies, including promoting female managers and encouraging managers to interact with all employees equally, to reduce the prevalence and effects of implicit bias.