Empirical insights on perceived fairness violation in algorithm-supported managerial decisions

People analytics – algorithmic management systems in personnel management – constantly collect and process employee-generated data. This can allow the systems to deliver actionable insights to support managerial decisions. Thus, people analytics can become a mediator between managers and employees, challenging and redefining their dynamics, relationships, and communication. In this study, we conduct an online experiment where participants take the role of employees, performing a real-effort task. In manipulated scenarios, rewards for completing the task are distributed either by a team leader without algorithmic support, a team leader with the option of algorithmic support, or a team leader who delegates the process of rewarding employees to an autonomous system. Our results indicate that the involvement of people analytics has downstream consequences, inducing feelings of unfairness and betrayal, which ultimately increase demands for reparation and general retaliatory behaviour against the manager and the organisation. Specifically, our findings show that, in unfair situations, employees perceive a stronger violation of fairness when they suspect a people analytics system to be involved in the decision-making process, and less violation when the same unfair decision was made by a manager without algorithmic support. Managers should therefore ensure that, even though they increasingly interact with technical systems, they keep people in focus.