Do managers who hire workers develop biases?

In new research to be presented at the annual conference of the Royal Economic Society, Alan Benson and Louis-Pierre Lepage examine whether managers risk developing biases and preconceptions through their hiring decisions.

When racial minorities apply for jobs, give birth, or interact with the judicial system, outcomes can depend on the individual bias of an influential decision-maker. In such settings, where the evaluator and evaluated are randomly paired with each other, potentially life-altering decisions can depend in part on the luck of the draw.

Institutions and organizations are making considerable efforts to de-bias decision-making in high-stakes settings, from reliance on diverse decision-making bodies, algorithmic recommendations, or implicit bias training. However, whether these solutions are effective depends on whether decision-makers’ beliefs are malleable, or whether they are fixed by the time they arrive on the job.

Using data on the sequence of hires at a large US retail chain, the authors find evidence that managers may increasingly discriminate “on the job”: negative experiences hiring black workers make them especially less likely to hire black workers in the future. And since these managers rarely hire black workers in the future, their biased beliefs about black workers are far slower to correct than for white workers. The results suggest biases and preconceptions are not fully formed by the time managers arrive on the job, but change based on their own, recent personal experience.

In their paper “Learning to Discriminate on the Job”, the authors find that biased beliefs are shaped by immediate personal experience on the job rather than immutable preconceptions, bias, or statistical evidence.

In reaching this conclusion, the authors examined the sequence of hiring decisions made by 27,470 managers at over 4,000 retail stores in the US. Retail managers in this setting have considerable autonomy when hiring for their departments; even within the same stores, there are significant differences in the race of new hires, seemingly based on the manager alone.

The authors investigated whether these differences arise in part from managers’ own previous experiences hiring workers of different races. For instance, if a manager hires a worker into a permanent position, and that worker turns over within two months, the authors count that as a negative experience and examine how it affects the propensity of the manager to hire workers of that race in the future. Short tenures are also highly correlated with poorer sales performance, terminations for poor performance, and other markers of a negative experience.

The authors find that negative experiences hiring black workers result in manager-specific declines in the probability that they hire black workers in the future. This pattern appears to be driven by a few factors.

First, managers seem to update their beliefs – both positively and negatively – about the performance of black workers more than they do for white workers. One possibility is that managers have less personal experience to draw upon when making decisions about black workers, so each instance is more salient.

Second, a manager’s positive or negative experiences with black or white workers change their hiring in the future, and they are slower to return to normal following negative experiences with black workers. Intuitively, this is because such negative experiences mean managers will develop a negative perception of black workers and rarely hire them, so they will be slow to gain the experience necessary to correct their own biases.

These two factors combine so that individual biases against black workers, and possibly others, are larger and more persistent.

The findings have important implications for our understanding of the emergence of biases, which appear moulded by specific individual life experiences, and to evolve over time rather than being predetermined by the time an employer begins their hiring career.

Minorities are inherently disadvantaged by the fact that managers stereotype them based on the performance of a few individuals they observe, leading to more negative perceptions of the group. The study also suggests that interventions that diffuse information and decision-making authority, such as algorithmic recommendations and hiring committees, might overcome individual manager biases and could yield greater diversity and better hires.


Alan Benson (University of Minnesota) – bensona@umn.edu, Twitter: @alanmbenson

Louis-Pierre Lepage (SOFI at Stockholm University) – louis-pierre.lepage@sofi.su.se

Notes to Editors:

The press release is highlighting research papers presented at the RES Annual Conference 2022 (#RES20220) for further information, please contact j.randalledwards@res.org.uk on 07970 201456 if you want the link to the full paper.

The Royal Economic Society’s popular 2022 virtual annual conference with in-person conference hubs, provides a forum for research, debate, and networking. The RES provides resources for economists and support for education and the training of students, teachers and researchers. 

The Royal Economic Society’s purpose is to promote the study of economic science. With over 6,500 members worldwide, RES is one of the oldest economic associations in the world. RES are a registered charity and membership is open to anyone who shares our aims and objectives.

The Society’s two peer-reviewed journals – The Economic Journal and The Econometrics Journal – contain high quality papers from an international authorship. 

Please do tag RES on Twitter @RoyalEconSoc using #RES2022

Or further RES Social Media pages: Website  YouTube  LinkedIn