The effects of leader power and status on employees' voice behavior: The role of psychological safety

The impact of a leader's hierarchical position on employees' voice behaviors is both practically and theoretically important. Prior research found that hierarchical differences hinder upward communication and information sharing. Indeed, the hierarchical position is constructed on different bases. For example, power (i.e., the control over valuable resources) and status (i.e., the respect and esteem that a leader holds in the eyes of others) are two different bases of hierarchy, and they are found to have distinct effects on an individual's behaviors and perceptions. Therefore, employees' voice decisions may vary when the leader has high power or high status. However, prior research has not distinguished those different hierarchical bases (i.e., power and status) when investigating the effect of a leader's hierarchical rank on employees' voice behaviors. Moreover, the interactive relationship between power and status has not been explored in the voice context, although researchers have identified that the power effect often depends on status. We suggest that leader power and status have distinct effects on employees' voice behavior and psychological safety. We also expect the effect of leader power on employee's voice behavior to be moderated by leader status and that the moderating effect may be transferred indirectly through employees' psychological safety. By taking a multimethod approach, we tested our hypotheses in two experiments (Studies 1 and 2) and a field survey (Study 3). In Study 1, we conducted a 2 (leader power: high vs. low) x 2 (leader status: high vs. low) between-subjects design and recruited 163 full-time employees from Prolific. By using a scenario-based experiment, we aimed to test the main effects of leader power and status on employees' voice behaviors, as well as the moderating effect of leader status. In Study 2, we conducted a 2 (leader power: high vs. low) x 2 (leader status: high vs. low) between-subjects design and recruited 189 full-time employees from Prolific. In this study, we asked participants to identify a target leader who had either high or low power and either high or low status. After that, they indicated how many suggestions or concerns they would bring up to this leader in a real meeting and reported the perceived psychological safety in that situation. In Study 3, we collected data from a large company in South China. The final sample consists of 346 employees matched with their 111 immediate supervisors. Employees evaluated their supervisor's power and status at T1 and reported their psychological safety at T2; supervisors reported employee voice behaviors at T2. In Study 1, we found a negative effect of leader power on employees' voice behaviors. In addition, we found that the effect of leader power on employees' voice behavior was contingent on leader status. That is, when a leader has high status, the negative effect of leader power on employee voice behavior was not significant; when a leader has low status, the negative effect was stronger. Study 2 supported the main effect of leader power and status on employees' voice behaviors, as well as the moderating effect of leader status. In addition, Study 2 provided evidence of the mediating role of psychological safety that links the interaction of leader power and status with employees' voice behaviors. Finally, Study 3 replicated the main effect of leader power and status on employees' voice behavior, the moderating effect of status on the link between leader power and employees' voice behavior, and the mediating effect of psychological safety. The results from two experiments and one field study ensured high internal and external validity. Our study contributes to the literature in several aspects. First, we distinguished the effect of leader power and status on employees' voice behaviors and psychological safety. We also found that the effect of leader power on employees' voice behavior depends on leader status. This finding provided more nuanced evidence regarding how hierarchical differentiation influences upward voice. Second, although researchers have identified leader power as an influential predictor of employees' voice behavior, the findings are mixed. We demonstrated that the impact of leader power is contingent on leader status. This finding helps reconcile the debate on leader power. Third, researchers have focused more attention on how leaders influence employees through hard power than on the influence of soft status. Our results show that dominance (i.e., power) and prestige (i.e., status) are both important sources of influence in organizations.