INTRODUCTION
As business environments become increasingly complex and uncertain, organizations continue to decentralize, gradually shifting the hierarchical focus of leadership towards a more collective and participative style (Yammarino, Salas, Serban, Shirreffs, & Shuffler, Reference Yammarino, Salas, Serban, Shirreffs and Shuffler2012). Despite this trend, researchers have found that authoritarian leadership, which emphasizes top-down communication and tight control (Farh & Cheng, Reference Farh, Cheng, Li, Tsui and Weldon2000), remains common in China (Chen, Li, & Leung, Reference Chen, Li and Leung2017; Takeuchi, Wang, & Farh, Reference Takeuchi, Wang and Farh2020) and other cultures (De Hoogh & Den Hartog, Reference De Hoogh and Den Hartog2009; Pellegrini & Scandura, Reference Pellegrini and Scandura2008). Many researchers have paid considerable attention to understanding this phenomenon by examining the function of authoritarian leadership. Unsurprisingly, evidence shows that it is negatively associated with subordinates’ outcomes, including organizational commitment, organizational citizenship behavior, and job performance (Chan, Huang, Snape, & Lam, Reference Chan, Huang, Snape and Lam2013; Cheng, Huang, & Chou, Reference Cheng, Huang and Chou2002; Schaubroeck, Shen, & Chong, Reference Schaubroeck, Shen and Chong2017; Wu, Huang, Li, & Liu, Reference Wu, Huang, Li and Liu2012).
However, mixed findings do exist. Encouraged by a weak positive relationship between authoritarian leadership and subordinate performance (Cheng, Chou, Huang, Farh, & Peng, Reference Cheng, Chou, Huang, Farh and Peng2003; Chou, Chou, Cheng, & Jen, Reference Chou, Chou, Cheng and Jen2010), researchers explore the persistence of authoritarian leadership from the functional view of hierarchy (Halevy, Chou, & Galinsky, Reference Halevy, Chou and Galinsky2011; Ronay, Greenaway, Anicich, & Gallinsky, Reference Ronay, Greenaway, Anicich and Galinsky2012). This perspective suggests that authoritarian leadership can be beneficial in pursuing operational efficiency and collective success in certain situations. For example, Huang, Xu, Chiu, Lam, and Farh (Reference Huang, Xu, Chiu, Lam and Farh2015) found that it was positively related to firm performance when the firm faced resource constraints. Leung, Chen, Zhou, and Lim (Reference Leung, Chen, Zhou and Lim2014) reported that group innovation performance was highest when group innovative climate and authoritarian leadership were both high. In a recent study, Wang et al. (Reference Wang, Tsai, Dionne, Yammarino, Spain, Ling, Huang, Chou and Cheng2018) found that when combined with leader benevolence, authoritarian leadership generated a positive effect on subordinate performance. In light of these findings, what, if any, are the risks of accepting the positive side of authoritarian leadership?
With the above question in mind, we suggest that when examining the performance implications of authoritarian leadership, researchers assign less attention to investigating its ethical consequences. The lack of this link is unfortunate, because we may mistakenly promote this leadership style at the cost of ethical standards. To demonstrate this possibility, we build and test a model linking authoritarian leadership and unethical pro-organizational behaviors (UPB) in this study. UPB serves our purpose well because it refers to ‘actions that are intended to promote the effective functioning of the organization or its members and violate core societal values, mores, laws, or standards of proper conduct’ (Umphress & Bingham, Reference Umphress and Bingham2011: 622). This examination thus focuses on the inherent tension between performance goals and ethical standards (Fehr et al., Reference Fehr, Welsh, Yam, Baer, Wei and Vaulont2019).
We employ the social cognitive theory of morality (Bandura, Barbaranelli, Caprara, & Pastorelli, Reference Bandura, Barbaranelli, Caprara and Pastorelli1996) to develop our hypotheses. This theory proposes that, in everyday life, individuals monitor their behaviors by comparing them to internal moral standards through self-regulatory processes. However, when they can use various explanations to exempt unethical behaviors from moral self-sanctions, namely moral disengagement, they may engage in such behaviors in the name of collective benefit (Bandura et al., Reference Bandura, Barbaranelli, Caprara and Pastorelli1996; Dang, Umphress, & Mitchell, Reference Dang, Umphress and Mitchell2017). Different from previous literature which often treats moral disengagement as a single, overarching mechanism underlying UPB (Chen, Chen, & Sheldon, Reference Chen, Chen and Sheldon2016; Umphress & Bingham, Reference Umphress and Bingham2011), we suggest displacement of responsibility, one component of moral disengagement, as the primary mechanism for understanding the relationship between authoritarian leadership and UPB. When individuals perceive that they cannot negotiate with the leader regarding their performance goal, they tend to interpret their unethical behaviors as the direct result of authoritative dictates. Equipped with such a shield of accountability, they are psychologically free to engage in UPB without moral self-sanctions (Umphress, Bingham, & Mitchell, Reference Umphress, Bingham and Mitchell2010). Meanwhile, the extent an individual's moral self-sanctioning is deactivated depends on not only the leader factor, but also the individual's insistence on moral standards in that context (Detert, Treviño, & Sweitzer, Reference Detert, Treviño and Sweitzer2008; Moore & Gino, Reference Moore and Gino2015). Deriving from such a personal-situation interactionist perspective (Terborg, Reference Terborg1981; Treviño, Reference Trevino1986), we expect that in the face of an authoritarian leader, an individual's moral self-regulation function can be protected by their commitment to maintaining moral performance (i.e., moral efficacy, Hannah & Avolio, Reference Hannah and Avolio2010). We thus examine moral efficacy as an important moderator in understanding the indirect relationship between authoritarian leadership and UPB through displacement of responsibility.
We make three contributions to the literature. First, by examining its ethical downsides, this study paints a more complete picture of the effects of authoritarian leadership. Previous research has largely focused on the influence of authoritarian leadership on employee performance outcomes but neglected to consider its ethical implications. This tendency has created great ambiguity in understanding its effect, especially given the accumulation of evidence for the positive effects of this leadership style. Second, we shed light on how displacement of responsibility is the most pertinent mechanism among moral disengagement components in the link between authoritarian leadership and UPB. Previous research often treated moral disengagement as a single overarching construct to explain UPB (Chen et al., Reference Chen, Chen and Sheldon2016; Umphress & Bingham, Reference Umphress and Bingham2011). However, moral disengagement embraces a set of different cognitive mechanisms. Treating moral disengagement as an overall construct would blur our understanding of an individual's self-regulatory process, particularly when we need to answer specific theoretical questions. As a response to the call ‘to study individual mechanisms of moral disengagement’, (Moore, Reference Moore2015: 202), our exploration should advance the understanding of how an individual mechanism of moral disengagement operates in specific contexts. Third, by examining moral efficacy as a moderating factor, the present investigation contributes to our knowledge of self-regulatory processes in ethical decision making. Researchers have urged for a deeper understanding of the factors and conditions that can facilitate or inhibit moral disengagement (Detert et al., Reference Detert, Treviño and Sweitzer2008; Shepherd, Patzelt, & Baron, Reference Shepherd, Patzelt and Baron2013). This study directly addresses this need and illustrates that an individual's positive moral experience can buffer the destructive effect of authoritarian leadership on moral self-regulation.
THEORETICAL BACKGROUND AND HYPOTHESES DEVELOPMENT
Recently UPB received a great deal of research attention, because these behaviors sacrifice ethical standards for short-term benefits (i.e., mispresenting the truth to sell the organization's products), and may damage the organization's reputation and customers’ trust in the long term (Umphress & Bingham, Reference Umphress and Bingham2011). Researchers largely rely on the social cognitive theory of morality to understand why such bad behaviors happen: when moral disengagement is activated under certain situations, individuals feel free from self-sanctions and the accompanying guilt when their behaviors violate internal moral standards (Chen et al., Reference Chen, Chen and Sheldon2016; Detert et al., Reference Detert, Treviño and Sweitzer2008; Umphress & Bingham, Reference Umphress and Bingham2011).
However, moral disengagement is an umbrella construct, which includes three main categories of tactics (Bandura et al., Reference Bandura, Barbaranelli, Caprara and Pastorelli1996). The first category of tactics includes reconstructing the moral implication of the conduct itself, including moral justification, euphemistic language, and palliative comparisons. The second category of tactics includes disclaiming the agentive role of the actor with displacement and diffusion of responsibility, as well as minimizing, ignoring, or misconstruing the consequences of their actions. The final category of tactics is used to dehumanize the victim or reduce any perceived distress caused by the unethical action to its victims. Considering their substantial differences, moral disengagement is best understood to be ‘multifaceted’, but not multifactorial (Bandura et al., Reference Bandura, Barbaranelli, Caprara and Pastorelli1996: 367). When individuals must find an appropriate way to construe their actions as morally justifiable, those tactics should be selectively employed (Moore & Gino, Reference Moore and Gino2015). For example, Kish-Gephart, Detert, Treviño, Baker, and Martin (Reference Kish-Gephart, Detert, Treviño, Baker and Martin2014) used attribution of blame and distortion of consequences because these two types of mechanisms were most likely to be evoked in their personal gain scenarios. Huang, Greenbaum, Bonner, and Wang (Reference Huang, Greenbaum, Bonner and Wang2019) assessed devaluation of target to capture employees’ rationalizations for customer-directed sabotage when they experienced customer mistreatment. Thus, specific situations may activate specific facets of moral disengagement that align with the context. Based on this understanding, displacement of responsibility is chosen as the specific mechanism to understand the indirect relationship between authoritarian leadership and UPB. This component refers to the tendency that an individual views their unethical behaviors and detrimental consequences as springing from authoritarian dictates (Hinrichs, Wang, Hinrichs, & Romero, Reference Hinrichs, Wang, Hinrichs and Romero2012), thus it is most likely to be activated among moral disengagement mechanisms when ‘an individual feels compelled to follow the orders of authority figures’ (Detert et al., Reference Detert, Treviño and Sweitzer2008: 376).
The Indirect Role of Displacement of Responsibility
Authoritarian leadership describes a leadership style that emphasizes rigorous control over subordinates and complete obedience from them (Farh & Cheng, Reference Farh, Cheng, Li, Tsui and Weldon2000). Due to the long experience of bitterness and difficulties, Chinese have a strong preference for social order over social chaos (Gabrenya & Hwang, Reference Gabrenya, Hwang and Bond1996). To maintain group efficiency and increase collective welfare, leaders are traditionally endowed with absolute power to make all important decisions and secure employee conformity with high requirements through the determination of rewards and punishments (Farh & Cheng, Reference Farh, Cheng, Li, Tsui and Weldon2000).
Rooted in the traditions of familism and paternalistic control in Chinese society (Westwood, Reference Westwood1997), the power of authoritarian leaders comes from the perception that subordinates are strongly dependent on their superiors’ abilities and experience, and thus their primary role is to convey the methods by which they have achieved success to subordinates and train them to achieve high levels of performance (Chen et al., Reference Chen, Li and Leung2017; Hamilton, Reference Hamilton1990). Consistent with this logic, subordinates should not only deliberately suppress their own ideas and opinions, but also do their best to accomplish tasks assigned by the leader without any questions (Chen et al., Reference Chen, Li and Leung2017; Farh & Cheng, Reference Farh, Cheng, Li, Tsui and Weldon2000). Therefore, when working with an authoritarian leader, individuals feel that there are fewer opportunities to express personal views to the leader and that they have limited capacity to influence the decision-making processes (Chan, Reference Chan2014; Tost, Gino, & Larrick, Reference Tost, Gino and Larrick2013). Consequently, their personal agency is strongly inhibited and the only thing they can do is achieve the assigned performance goal set by the leader. Viewing themselves as just an extension of the authority, it is difficult for individuals to activate moral self-sanctions to judge their own actions that are employed to meet the leader's requirement (Butterfield, Treviño, & Weaver, Reference Butterfield, Trevin and Weaver2000; Treviño, Reference Trevino1986). Thus, the implementation of authoritarian leadership blurs the association between individuals’ own actions and moral consequences and enables them to view their choices as stemming from the leader's orders (Bandura, Reference Bandura2016).
When displacement of responsibility is activated by the leader's dominating behaviors, UPB, such as exaggerating the truth about company's products to customers and clients, becomes a convenient approach to fulfilling leader's high requirements and expectations. As the representative of an organization, the leader usually has the authority to make decisions that strongly influence subordinates (e.g., promotion, pay, or work assignment) (Whitener, Brodt, Korsgaard, & Werner, Reference Whitener, Brodt, Korsgaard and Werner1998). Thus, subordinates realize that to access valued resources and developmental opportunities, they have to impress the leader and achieve the high-performance goals set by the authoritarian leader. They have no other choices because there is less leeway to negotiate with the leader about their own goals (Chan, Reference Chan2014; Farh & Cheng, Reference Farh, Cheng, Li, Tsui and Weldon2000). Different from other compliance behaviors, the essence of UPB remains unethical and violates moral standards (Umphress & Bingham, Reference Umphress and Bingham2011). However, whenever displacement of responsibility is activated, UPB becomes highly likely to occur as individuals can justify their unethical actions by describing them as merely achieving the goal set by the authority, and thus free themselves from moral self-sanction and the effects of deterrent mechanisms (e.g., guilt) regarding unethical behaviors (Hinrichs et al., Reference Hinrichs, Wang, Hinrichs and Romero2012). As a result of these processes, the relationship between UPB and personal accountability is blurred, and the perceived moral intensity of UPB is decreased (Barsky, Reference Barsky2011). For them, UPB now becomes a productive way to reach the assigned goal set by the leader. They believe that the leader should be responsible for the moral consequences and that they are a ‘mere functionary obeying orders’ (Bandura, Reference Bandura2016: 59). Hence, we propose the following hypotheses:
Hypothesis 1a: Authoritarian leadership is positively related to an individual's displacement of responsibility.
Hypothesis 1b: There is a significant indirect relationship between authoritarian leadership and a subordinate's UPB via displacement of responsibility.
The Moderating Role of Moral Efficacy
The above logic suggests that the restriction of autonomy and personal agency based on authoritarian leadership blurs the association between subordinates’ UPB and their moral consequences, thus triggering the displacement of responsibility mechanism. However, social cognitive theory of morality suggests that the strength of moral self-sanctioning is shaped by personal moral standards (Detert et al., Reference Detert, Treviño and Sweitzer2008; Moore & Gino, Reference Moore and Gino2015). Focusing this logic to the current context, we propose that moral efficacy should be able to influence to what extent authoritarian leadership evokes displacement of responsibility. Moral efficacy refers to an individual's belief that they have the capability to mobilize cognitive resources to attain moral performance when facing an ethical dilemma (Hannah & Avolio, Reference Hannah and Avolio2010). When an individual has high moral efficacy, a rich personal mastery, and vicarious experiences in moral performance help them to set high moral standards and display high self-directedness (Hannah, Avolio, & May, Reference Hannah, Avolio and May2011; May, Luth, & Schwoerer, Reference May, Luth and Schwoerer2014; Nelson, Poms, & Wolf, Reference Nelson, Poms and Wolf2012). Thus, when working with an authoritarian leader, they tend to believe that they are capable of achieving moral purpose even when they are in a relatively powerless situation. Their cognitive resources help them establish such confidence and mobilize their agency to take their own moral stance (Lee, Choi, Youn, & Chun, Reference Lee, Choi, Youn and Chun2017; Zimmerman & Kitsantas, Reference Zimmerman and Kitsantas2005). Hence, in the face of an authoritarian leader, their inner moral compass makes them less likely shirk the responsibilities of their own behaviors to the leader. This possibility weakens the relationship between authoritarian leadership and displacement of responsibility.
In contrast, an individual with low moral efficacy has neither the confidence nor the cognitive resources to insist on a high moral standard in the face of contextual pressures. Facing the leader's strict discipline and high demands, the powerlessness feeling makes their self-monitoring mechanisms even more fragile. With a strong motivation to achieve the leader's assigned goal, their moral standards fall easily and the corresponding self-sanctioning functions are likely to be deactivated, enabling them to easily shirk the responsibility for their own actions (Fida, Tramontano, Paciello, Ghezzi, & Barbaranelli, Reference Fida, Tramontano, Paciello, Ghezzi and Barbaranelli2016; Welsh & Ordóñez, Reference Welsh and Ordóñez2014). Consequently, the relationship between authoritarian leadership and displacement of responsibility would be enhanced. We thus conclude that when the environment (i.e., an authoritarian leader) provides opposition or pressure that conflicts with an individual's moral standards, their efficacy belief plays a buffering role in exercising personal agency and taking moral responsibility. Thus, we hypothesize:
Hypothesis 2a: Moral efficacy negatively moderates the relationship between authoritarian leadership and displacement of responsibility, such that the positive relationship is stronger when moral efficacy is lower rather than when it is higher.
Based on Hypothesis 1 and 2a, we propose a conditional indirect relationship: the indirect relationship between authoritarian leadership and UPB via displacement of responsibility is contingent on an individual's moral efficacy. When they face an authoritarian leader, individuals with high moral efficacy have the confidence that they can meet the leader's requirements while adhering to their own ethical standards. Hence, they are likely to take moral responsibility for their actions and engage in fewer UPB. In contrast, when an individual with low moral efficacy works with an authoritarian leader, they are likely to activate the mechanism of displacement of responsibility which leads to more UPB. Hence:
Hypothesis 2b: Moral efficacy negatively moderates the indirect relationship between authoritarian leadership and UPB via displacement of responsibility, such that this indirect relationship is stronger when moral efficacy is lower rather than when it is higher.
OVERVIEW OF STUDIES
We progressively tested our hypotheses in two studies. In Study 1, we collected data using a vignette-based experiment to examine the indirect relationship between authoritarian leadership and UPB via displacement of responsibility (Hypotheses 1a and 1b). In addition, to establish strong inferences about the indirect role of displacement of responsibility, we tested its effects over those of other moral disengagement components, which may be activated in the authoritarian leadership context. To test the full research model and replicate findings in Study 1, we further conducted a multisource survey-based field study (Study 2). In this study, we used a different version of the displacement of responsibility scale, and changed the data source of UPB from self-reports to supervisory ratings.
STUDY 1
Participants, Task, and Manipulation
The purpose of this study is to test the indirect relationship between authoritarian leadership and UPB via displacement of responsibility. Built on Cheng and colleagues’ conceptualization and operationalization (Cheng, Chou, & Farh, Reference Cheng, Chou and Farh2000; Farh & Cheng, Reference Farh, Cheng, Li, Tsui and Weldon2000) and our interviews in a software company, we developed two scenarios to manipulate authoritarian leadership (high vs. low). Two hundred and twenty-seven participants were recruited and randomly assigned to the two conditions with the help of an online-marketing company. Before being invited to participate in our study, a pre-screening questionnaire was designed to check the qualifications of the participants (i.e., must be full-time employees with at least 2 years of working experience). Following the recommendations of Meade and Craig (Reference Meade and Craig2012), we included three attention check questions and excluded participants who responded carelessly. At the end, 198 participating results were retained, representing a response rate of 87.2%. The final sample contained 95 women (48.0%), mostly aged 26~30 years old. Eighteen participants held a masters’ degree or above (9.1%), 158 participants had a bachelors’ degree (79.8%), and 22 participants had educational experience in a professional school or high school (11.1%).
Participants were randomly assigned into the two scenarios (see details in the Appendix) and were asked to imagine that they were a salesman in a marketing department of a software development company. In the high authoritarian leadership scenario (n = 100), they learned that their supervisor, Liu Yang, always imposes paternalistic control at work. He requires the subordinates to adhere to the high sales goals he made and never negotiates such assignments with them. In the low authoritarian leadership scenario (n = 98), they learned that their supervisor always considers employees’ opinions before making their sale goals and behaves kindly and openly to employees. After this part, all participants were presented with a business situation in which they needed to sell logistic management software with certain unnoticed bugs. Last, they reported their level of displacement of responsibility and their intention of promoting this software to the customer.
Measures
A five-point Likert scale was used to measure participants’ responses (1= strongly disagree, 5 = strongly agree). All materials were presented in Chinese. English items were translated into Chinese following translation and back-translation procedures (Brislin, Reference Brislin, Triandis and Berry1980).
Intention to engage UPB
After reading the scenario, participants were asked to report their intention to engage in UPB. Four items developed by Umphress et al. (Reference Umphress, Bingham and Mitchell2010) were adapted to this research context. Sample items included ‘to help my organization increase its market share, I would exaggerate the truth about this software to customers and clients’ and ‘to increase my organization's sale performance, I would withhold negative information about the software to the clients’. Its Cronbach's α was 0.74.
Displacement of responsibility
In the literature, displacement of responsibility was measured as either an individual difference variable (Moore, Detert, Treviño, Baker, & Mayer, Reference Moore, Detert, Treviño, Baker and Mayer2012) or a cognitive mechanism that varies across different situations (Moore, Reference Moore2015). Consistent with the recent contention that the leader is a key influence on the extent to which subordinates actively disengages morally cognitions (Moore et al., Reference Moore, Mayer, Chiang, Crossley, Karlesky and Birtch2019), we adapted the three items developed in Moore et al. (Reference Moore, Detert, Treviño, Baker and Mayer2012) to measure displacement of responsibility as a cognitive mechanism in this study. A sample item was ‘I might not be held accountable for doing questionable things because my boss told me to accomplish the task’. Its Cronbach's α was 0.80.
Control variables
To generate a clear conclusion regarding the hypothesized indirect relationship, we controlled three categories of variables: First, we included demographic variables, including age, gender, and education, to account for their influence on UPB (Kish-Gephart, Harrison, & Treviño, Reference Kish-Gephart, Harrison and Treviño2010; Umphress et al., Reference Umphress, Bingham and Mitchell2010). Specifically, we measured gender as a dummy variable coded as 1 for female and 0 for male, age by using five continuous categories (20 or below, 21–30, 31–40, 41–50, and 51 or above), and education was measured using categories (high school or below, college, university, and postgraduate).
Second, authoritarian leadership may result in a subordinate's UPB by inducing fear of the leader (Cheng, Chou, Wu, Huang, & Farh, Reference Cheng, Chou, Wu, Huang and Farh2004). This possibility suggests that an individual may not change their moral cognitions in ethical decision making. To eliminate this possibility, we included the three-item scale of fear of the leader used by Cheng et al. (Reference Cheng, Chou, Wu, Huang and Farh2004). A sample item is ‘I'm afraid of Liu Yang’. Its Cronbach's α was 0.88.
Last, we controlled for other possible moral disengagement components in our hypothesized context. As reviewed, moral disengagement includes multiple cognitive mechanisms, which have the potential to be mediators as well. For example, authoritarian leadership may relate to diffusion of responsibility, because under the leader's high demand and tight control, every team member may be doing the same thing. Meanwhile, authoritarian leadership may lead to high moral justification (i.e., justifying the behavior by a high-end goal) and attribution of blame (i.e., using mistreatment from the leader to justify unethical behaviors). To rule out such alternative explanations, we measured these mechanisms by adapting items from Moore et al.'s (Reference Moore, Detert, Treviño, Baker and Mayer2012) scale. Sample items included ‘It is okay for me to tell a lie if everyone agrees that it's the best way to handle the situation’ (diffusion of responsibility, Cronbach's α = 0.77), ‘It is okay for me to withhold potentially damaging information to protect the company's interests’ (moral justification, Cronbach's α = 0.82), and ‘If the client buys this product, it's probably because he has not taken adequate precautions to protect his interest’ (attribution of blame, Cronbach's α = 0.72).
Results
Manipulation check
We used three items from Cheng et al.'s (Reference Cheng, Chou, Wu, Huang and Farh2004) authoritarian leadership scale to check the validity of our manipulation. A sample item was ‘Liu Yang has asked us to obey his/her instructions completely’. Its Cronbach's alpha was 0.96. Participants in the high condition reported a significantly higher score of authoritarian leadership (M = 4.52, SD = 0.73) than participants in the low condition (M = 1.74, SD = 1.03; t(196) = 21.93, p < 0.01). Thus, this manipulation was perceived as intended by the participants.
Descriptive statistics
Table 1 presents the means, standard deviations, internal consistency reliability coefficients, and correlations of the variables.
Table 1. Means, standard deviations, correlations, and internal consistency estimates (study 1)
Notes: N = 198; Standardized internal consistency reliability estimates (Cronbach's α) appear in parentheses along the diagonal. *p < 0.05. **p < 0.01.
Hypotheses testing
In support of Hypothesis 1a, the mean displacement of responsibility value in the high authoritarian leadership condition was 3.13 (SD = 1.02) and 2.58 (SD = 0.81) in the low condition. A two-sample t test indicated significant difference (t(196) = - 4.18, p < 0.01). Similarly, the mean fear of the leader values for the high (M = 3.76, SD = 0.88) and low (M = 2.09, SD = 0.88) conditions were significantly different (t(196) = -13.33, p < 0.01). However, participants did not report significantly different values for attribution of blame (Mhigh = 2.81, SDhigh = 0.91; Mlow = 2.66, SDlow = 0.86; t(196) = -1.27, n.s.), moral justification (Mhigh= 2.47, SDhigh = 0.96; Mlow = 2.29, SDlow = 1.08; t(196) = -1.27, n.s.) and diffusion of responsibility (Mhigh = 2.79, SDhigh = 0.90; Mlow = 2.63, SDlow = 0.78; t(196) = -1.37, n.s.). Taken together, Hypothesis 1a was support.
Hypothesis 1b proposes the indirect relationship of authoritarian leadership with UPB through displacement of responsibility. We tested this hypothesis using the PROCESS SPSS macro (Hayes, Reference Hayes2012) and included demographic variables as covariates. A 5000-sample bootstrapping procedure was used to generate bias-corrected confidence intervals (CIs). As presented in Figure 1, authoritarian leadership had a significant positive relationship with displacement of responsibility (B = 0.54, s.e. = 0.13, p < 0.01), and displacement of responsibility had a significant relationship with UPB intention (B = 0.15, s.e. = 0.06, p < 0.05). Furthermore, the indirect relationship between authoritarian leadership and UPB via displacement of responsibility was significant (B = 0.08, s.e. = 0.04, 95% bias-corrected CI = [0.01, 0.18], excluding zero).
Figure 1. Summary of the proposed research model with possible mediating mechanisms (study 1)
Notes: Unstandardized coefficients are shown in the diagram and standard errors are included in brackets. * p<0.05; **p<0.01, two-tailed tests.
To rule out other alternative mechanisms, we also included fear of the leader, moral justification, attribution of blame, and diffusion of responsibility in the estimation. However, we did not find a significant indirect relationship between authoritarian leadership and UPB via attribution of blame (B = 0.01, s.e. = 0.02, 95% bias-corrected CI = [-0.02, 0.04], including zero), moral justification (B = 0.06, s.e. = 0.05, 95% bias-corrected CI = [-0.03, 0.10], including zero), diffusion of responsibility (B = 0.02, s.e.= 0.02, 95% bias-corrected CI = [-0.01, 0.09], including zero), or fear of leader (B = -0.14, s.e. = 0.09, 95% bias-corrected CI = [-0.35, 0.03], including zero). Thus, Hypothesis 1b was supported.
The above findings provided initial support for the indirect relationship specified in Hypotheses 1a and 1b above other alternative mechanisms. However, UPB was measured as self-reported behavioral intentions, not actual behaviors. Authoritarian leadership is artificial and may not be generalizable to real organizational settings. Therefore, we conducted Study 2 using a field sample and sought to capture the perceptions of real authoritarian leadership and employees’ UPB to replicate these findings and test the full research model.
STUDY 2
Sample and Procedures
We collected survey data from a large pharmaceutical retail company located in China. In focused group interviews conducted before data collection, we found that because of the intensifying competition in the retail market, our participants are motivated to improve their store's sales performance by engaging in unethical behaviors, such as exaggerating the curative properties of a certain medicine, withholding side effect information, and/or promoting unnecessary health care products to customers.
We collected data on site at two time points. All participants were assured that their responses were completely confidential and would only be used for research. At Time 1, participants were asked to describe their demographic information, and assess the items on authoritarian leadership and moral efficacy. After one month (Time 2), we asked them to report their level of displacement of responsibility. Meanwhile, we chose to collect each participant's UPB from the store manager. In the literature, researchers usually use self-reported UPB as we did in Study 1 (Chen et al., Reference Chen, Chen and Sheldon2016; Lee, Schwarz, Newman, & Legood, Reference Lee, Schwarz, Newman and Legood2017). However, this may be biased because individuals have the incentive to hide such information and are reluctant to report their UPB honestly. In the interview before data collection, store managers expressed their confidence in assessing each member's UPB, as the store size is not too large (ranging from 3–15, two shifting per day) and they are working in public areas with intensive interactions. In appreciation of their participation, we provided a small gift worth 15 RMB (approximately 2 dollars) to each respondent after data collection.
With the help of the HR manager and store managers, 290 subordinates and 35 store managers participated at Time 1. At Time 2, we received 202 usable surveys from subordinates from 35 stores, resulting in a response rate of 69.7% percent for subordinates. The final sample contained responses from 86 women (42.6%). Of the respondents, 21.3% had a middle school or below education, 58.9% had a college degree, 19.3% had a university degree, and 0.5% had a postgraduate degree. Most of them (70.4%) were below 40 years old.
Measures
We used 5-point Likert scales (1= strongly disagree, 5 = strongly agree) to measure participants’ responses. All English items were translated into Chinese following translation and back-translation procedures (Brislin, Reference Brislin, Triandis and Berry1980).
Authoritarian leadership
We measured this variable by the 9-item scale used in Cheng et al. (Reference Cheng, Chou, Wu, Huang and Farh2004). It has been widely used in the Chinese context with sound psychometric evidence (e.g., Schaubroeck et al., Reference Schaubroeck, Shen and Chong2017; Wu et al., Reference Wu, Huang, Li and Liu2012). A sample item was ‘My supervisor has asked me to obey her/his instructions completely’. Its Cronbach's alpha was 0.91.
UPB
We measured this variable using the six items adapted from Umphress et al. (Reference Umphress, Bingham and Mitchell2010). Based on the interview before data collection, we modified their items to fit the current context. A sample item was ‘To increase the store's sale performance, he/she withholds negative information of certain medicines from customers and clients’. Its Cronbach's alpha was 0.85.
Displacement of responsibility
We measured this variable by adapting three items from the scale of McFerran, Aquino, and Duffy (Reference McFerran, Aquino and Duffy2010). Different from Moore et al. (Reference Moore, Detert, Treviño, Baker and Mayer2012)'s scale we used in Study 1, this scale captures an individual's propensity to shift their own responsibility for unethical behaviors in the social context of workplace (McFerran et al., Reference McFerran, Aquino and Duffy2010). It has been used to capture an individual's self-regulation processes by other researchers in the Chinese context (e.g., Huang, Wellman, Ashford, Lee, & Wang, Reference Huang, Wellman, Ashford, Lee and Wang2017). A sample item was ‘If an employee is pressured into doing something, she/he shouldn't be blamed for it’. Its Cronbach's alpha was 0.84.
Moral efficacy
We measured it using the 5-item scale adapted from Hannah and Avolio (Reference Hannah and Avolio2010). A sample item was ‘I am confident that I can determine what needs to be done when I face a moral decision’. Its Cronbach's alpha was 0.81.
Control variables
Similar to Study 1, we included three demographic variables potentially related to the UPB: age, gender, and education. Specifically, we measured gender as a dummy variable coded as 1 for female and 0 for male, age by using five continuous categories (20 or below, 21–30, 31–40, 41–50, and 51 or above), and education was measured using categories (high school or below, college, university, and postgraduate).
Different from the fictitious context in Study 1, our participants worked within real stores. Thus, competitive pressure may motivate them to find an efficient way (i.e., UPB) to outperform others and obtain recognition and rewards (Chen, Zhu, & Zhou, Reference Chen, Zhu and Zhou2015; Kilduff, Galinsky, Gallo, & Reade, Reference Kilduff, Galinsky, Gallo and Reade2016). Thus, we controlled for perceived intra-team competition, which was measured by a four-item scale adapted from Fletcher, Major, and Davis (Reference Fletcher, Major and Davis2008). A sample item was ‘The amount of recognition I get in this company depends on how I perform compared to other members in our group’. Its Cronbach's alpha was 0.81.
Analytical Strategy
Given the nested structure of the data (i.e., a store manager provided assessments for several members), we employed multilevel modeling procedures via Mplus 7.0 (Muthén & Muthén, Reference Muthén and Muthén2012) to test our hypotheses. To account for non-normal sampling distributions of indirect relationships, we tested indirect and conditional indirect relationships using parameter-based resampling approach to calculate bias-corrected confidence intervals by using 20,000 resamples via R program (Preacher & Selig, Reference Preacher and Selig2012).
Results
Confirmatory Factor Analyses (CFA)
Before hypotheses testing, we conducted a series of CFA procedures to examine the discriminant validity of five measures: authoritarian leadership, displacement of responsibility, UPB, moral efficacy, and perceived intra-team competition. The data demonstrated a reasonable fit with the baseline five-factor model (χ2 = 260.16, d.f. = 160, RMSEA = 0.05, CFI = 0.95, SRMR = 0.06) and all factor loadings were significant. More importantly, this baseline model provided a better fit than alternative models, including a four-factor model with authoritarian leadership and displacement of responsibility collapsed into one factor (χ2 = 524.91, d.f. = 164, RMSEA = 0.10, CFI = 0.81, SRMR = 0.11), a four-factor model with displacement of responsibility and UPB collapsed into one factor (χ2 = 487.68, d.f. = 164, RMSEA = 0.10, CFI = 0.83, SRMR = 0.09), and the final four-factor model with moral efficacy and perceived intra-team competition collapsed into one factor (χ2 = 525.15, d.f. = 164, RMSEA = 0.10, CFI = 0.81, SRMR = 0.10).
Descriptive statistics
Table 2 presents the means, standard deviations, internal consistency reliability coefficients, and correlations of the variables in this study.
Table 2. Means, standard deviations, correlations, and internal consistency estimates (study 2)
Notes: N = 202; Standardized internal consistency reliability estimates (Cronbach's α) appear in parentheses along the diagonal. *p < 0.05. **p < 0.01.
Hypotheses testing
Hypotheses 1 proposed an indirect relationship between authoritarian leadership and UPB via displacement of responsibility. As shown in Model 1 of Table 3, after controlling for the effect of demographic variables, authoritarian leadership was positively related to displacement of responsibility (B = 0.27, s.e. = 0.08; p < 0.01). Thus, Hypothesis 1a was supported.
Table 3. Results of regression analyses with displacement of responsibility and UPB (study 2)
Notes: N = 202; Numbers in parentheses are standard errors of the corresponding parameter estimation. *p < 0.05. **p < 0.01.
As shown in Model 5 of Table 3, when authoritarian leadership was included in the model (B = -0.07, s.e. = 0.04; n.s.), displacement of responsibility was still positively related to UPB (B = 0.13, s.e. = 0.05; p < 0.01). However, the relationship between authoritarian leadership and UPB was not significant (B = 0.03, s.e. = 0.04, n.s., Model 4). We thus tested the indirect relationship between authoritarian leadership and UPB via displacement of responsibility by using Monte Carlo simulation, and found that this indirect relationship was significant (B = 0.03, s.e. = 0.02; 95% bias-corrected CI = [0.01, 0.06], excluding zero). Thus, Hypothesis 1b was supported.
Hypothesis 2a proposed that moral efficacy weakens the relationship between authoritarian leadership and displacement of responsibility. As shown in Model 3 in Table 3, the interaction term of authoritarian leadership and moral efficacy was negatively related to displacement of responsibility (B = -0.42, s.e. = 0.11, p < 0.01). We plotted this interaction effect in Figure 2. Following Preacher, Curran, and Bauer (Reference Preacher, Curran and Bauer2006), we further conducted simple slope tests and found that the relationship between authoritarian leadership and displacement of responsibility was positively significant in the low moral efficacy group (1SD below the mean, B = 0.54, s.e. = 0.10, p < 0.01). However, this relationship was not significant in the high moral efficacy group (1SD above the mean, B = 0.12, s.e. = 0.09, n.s.). The difference between the two groups was significant (B diff = -0.42, s.e. = 0.11, p < 0.01). Thus, Hypothesis 2a was supported.
Figure 2. Interaction of authoritarian leadership and moral efficacy on displacement of responsibility (study 2)
Continuing the above analyses, we tested whether the indirect relationship between authoritarian leadership and UPB via displacement of responsibility is moderated by moral efficacy (Hypothesis 2b). The results suggest that the indirect relationship between authoritarian leadership and UPB was significant in the low moral efficacy group (B = 0.06, s.e. = 0.03, 95% bias-corrected CI = [0.01, 0.12], excluding zero). Meanwhile, this indirect relationship was not significant in the high moral efficacy group (B = 0.01, s.e. = 0.01, 95% bias-corrected CI = [-0.01, 0.04], including zero). Importantly, the difference between the two groups was significant (Bdiff = -0.05, s.e. = 0.02, 95% bias-corrected CI = [-0.09, -0.01], excluding zero). Thus, Hypothesis 2b was supported.
DISCUSSION
Using a vignette-based experimental study (Study 1) and a multisource survey study (Study 2), we found consistent evidence that authoritarian leadership had an indirect positive relationship with UPB through displacement of responsibility. In Study 2, we found that this indirect relationship was stronger among employees with low moral efficacy. These findings offer meaningful implications for both future research and managerial practices.
Theoretical Implications
First, our research contributes to the understanding of authoritarian leadership by linking it with unethical employee outcomes. Research on authoritarian leadership has accumulated ample evidence suggesting that it is negatively related to employee work outcomes (Chan et al., Reference Chan, Huang, Snape and Lam2013; Cheng et al., Reference Cheng, Huang and Chou2002; Schaubroeck et al., Reference Schaubroeck, Shen and Chong2017; Wu et al., Reference Wu, Huang, Li and Liu2012). Despite this body of evidence, the positive side of authoritarian leadership has been reported in recent research (Huang et al., Reference Huang, Xu, Chiu, Lam and Farh2015; Leung et al., Reference Leung, Chen, Zhou and Lim2014; Wang et al., Reference Wang, Tsai, Dionne, Yammarino, Spain, Ling, Huang, Chou and Cheng2018). Derived from the functional view of hierarchy (Halevy et al., Reference Halevy, Chou and Galinsky2011), this line of research contends that authoritarian leadership may be uniquely effective in improving organizational operational efficiency and collective success in some situations. Regarding such mixed evidence, our findings suggest that although authoritarian leadership may not directly lead to UPB, it can deactivate an individual's self-regulatory processes and results in unethical behaviors in the name of the leader. Thus, it is highly possible that to impress the authoritarian leader, subordinates may use unethical methods to meet the assigned goal and achieve desirable outcomes. We should be cautious in highlighting the efficiency function of authoritarian leadership, because subordinates’ unethical actions may damage the company's reputation and performance in the long run.
Second, our findings demonstrate that displacement of responsibility can function as a specific cognitive mechanism to explain the relationship between authoritarian leadership and UPB. In the literature, moral disengagement is often treated as an overall construct to explain UPB (Chen et al., Reference Chen, Chen and Sheldon2016; Umphress & Bingham, Reference Umphress and Bingham2011). However, moral disengagement includes different facets and specific situational features often call for specific disengagement tactics (Kish-Gephart et al., Reference Kish-Gephart, Detert, Treviño, Baker and Martin2014). As a support for such arguments, our findings offer empirical evidence for the indirect relationship between authoritarian leadership and UPB via displacement of responsibility (both Study 1 and Study 2), while controlling for other possible moral disengagement components and fear of the leader (Study 1). Such knowledge regarding displacement of responsibility provides a more nuanced analysis of the effect of authoritarian leadership on UPB. We thus encourage researchers to explore the differences among moral disengagement components and achieve a clear understanding of the underlying mechanism behind certain ethical decisions or behaviors.
In addition, the present study also extends the research on how authoritarian leadership influences subordinates’ cognitive processes. Until now, researchers have focused on how authoritarian leadership influences subordinates’ role perception (Zhang & Xie, Reference Zhang and Xie2017), organization-based self-esteem (Chan et al., Reference Chan, Huang, Snape and Lam2013), perceived insider status (Schaubroeck et al., Reference Schaubroeck, Shen and Chong2017), and attitudes toward the leader such as trust in the leader (Chen, Eberly, Chiang, Farh, & Cheng, Reference Chen, Eberly, Chiang, Farh and Cheng2014). As an extension, our study suggests that authoritarian leadership can change an individual's self-regulation processes in ethical decision makings. This exploration of possible cognitive mechanisms clarifies how authoritarian leadership influences subordinates’ perceptions and beliefs at work.
Finally, our findings identified moral efficacy as an important boundary condition for the indirect relationship between authoritarian leadership and UPB through displacement of responsibility. Specifically, we found that moral efficacy significantly weakened the strength of this indirect relationship: individuals with lower levels of moral efficacy tend to display weaker self-directedness when they work with an authoritarian leader than their counterparts with higher levels of moral self-efficacy. This result is consistent with the predictions of the social cognitive theory of morality, which suggests that individuals vary in their self-monitoring capability and orientations (Detert et al., Reference Detert, Treviño and Sweitzer2008; Moore & Gino, Reference Moore and Gino2015). The moderating effect of moral efficacy not only provides supports for the possibility that the command-control approach leadership gives individuals an excuse to reduce cognitive dissonance resulting from their unethical behaviors, but also draws research attention to the role of individual differences in our understanding of moral self-regulation processes.
Practical Implications
Hierarchical relationships are a basic feature of business organizations. The duty to obey authority figures is widely supported in our moral norms and teachings (Beu & Buckley, Reference Beu and Buckley2004). Therefore, business leaders often have the incentive to impose high performance goals to their people, so to achieve organizational efficiency in market competition. However, our study suggests that authoritarian leadership may be an indirect source of employees’ UPB. For example, to improve sales performance, employees at Apple slowed iPhone processors by using aging batteries (Lee et al., Reference Lee, Schwarz, Newman and Legood2017). While helping companies to achieve the short-term performance goal, such unintended ethical outcomes may hurt companies’ reputation and bring serious consequences in the long run. Our leaders should be cautious about the ethical costs in pursuing organizational efficiency.
To avoid this situation, leaders should avoid demanding complete compliance from their subordinates, respect them, and seek their participation in collective decisions. Meanwhile, organizations should equip their people with the skills and capabilities necessary to properly analyze and determine their degree of involvement in potentially unethical behaviors. Employees need to be taught on how to determine whether a demand from their leader is legitimate and to be permitted to, without reprisal, seek further information when they encounter ethical dilemmas. When moral self-regulatory mechanisms are activated, managers can effectively facilitate ethical behaviors and inhibit unethical intentions. This, in the long-term, will aid the success of their organization.
Limitations and Future Research Directions
Several limitations in this study provide new insights for future research. First, our findings raise doubts regarding the efficiency logic in understanding the effect of authoritarian leadership: it is likely that an individual engages in UPB to achieve desirable outcomes and meet the authoritarian leader's expectations. However, we did not examine the consequences (i.e., task performance) when an individual sacrifices their moral standards and engage in UPB. Thus, future research is needed to explore the possible consequences of employee UPB and examine how such unethical outcomes bring negative consequences for the organization in the long run.
Second, even though two studies were conducted, the causal consequence between displacement of responsibility and UPB remains unclear. While Bandura et al. (Reference Bandura, Barbaranelli, Caprara and Pastorelli1996) argued that moral disengagement processes happen before unethical behaviors, it is likely that an individual would find explanations to avoid moral self-sanctions after engaging in unethical behaviors (Anand, Ashforth, & Joshi, Reference Anand, Ashforth and Joshi2004). Regarding this possibility, we cannot rigorously separate the two constructs and test their causal relationship. In addition, in Study 2, we only measured displacement of responsibility and did not control other moral disengagement mechanisms as we did in Study 1. Future research should address such limitations and obtain a clearer picture of an individual's cognitive process in engaging in UPB.
Third, in addition to displacement of responsibility, other alternative mechanisms may exist to explain the relationship between authoritarian leadership and UPB. For example, it is possible that under the tight control of authoritarian leaders, the team gradually develops a group-level instrumental norm about using particular practices (such as UPB) to improve performance. Then, employees may learn from their peers to engage in UPB as a result of socialization within the group. In addition, we should admit that competing mechanisms may exist to argue against the relationship between authoritarian leadership and UPB. For example, Jiang, Chen, Sun, and Yang (Reference Jiang, Chen, Sun and Yang2017) found that authoritarian leadership led to organizational cynicism from subordinates. As a consequence, people may become less commit to the organization and are less likely to engage in UPB. More research is needed to articulate the relationship between authoritarian leadership and UPB.
Fourth, we focused on authoritarian leadership only, without investigating the other two paternalistic leadership components, benevolent leadership (i.e., protecting the members and giving them what they need) and moral leadership (i.e., sticking to a high moral standard). Since all of these are intertwining components of the paternalistic approach to leadership (Farh & Cheng, Reference Farh, Cheng, Li, Tsui and Weldon2000; Wang et al., Reference Wang, Tsai, Dionne, Yammarino, Spain, Ling, Huang, Chou and Cheng2018), it is theoretically promising to include the other components and examine the ethical issues raised in this study. For example, our findings provide insights for understanding the relationship between authoritarian leadership and moral leadership. Even though a negative association was observed in the literature (see Hiller, Sin, Ponnapalli, & Ozgen, Reference Hiller, Sin, Ponnapalli and Ozgen2019, for a meta-analytic review), research that explains this negative relationship remains lacking. As an extension of our findings, we expect that attributing the responsibility for their unethical behaviors to the leader decreases followers’ moral evaluation of their leaders. Future research may explore this possibility and further examine the boundary conditions of the negative relationship. We believe such exploration should contribute to our understanding of how the different components of paternalistic leadership intertwine with each other and make an overall effect on subordinates.
The other possibility is to explore how the integration of authoritarian and benevolent leadership influences an individual's ethical decision making processes. In a recent study, Zheng, Huang, Graham, Redman, and Hu (Reference Zheng, Huang, Graham, Redman and Hu2020) found that authoritarian leadership deters employees’ deviance behaviors when leaders showed low benevolence and when employees were highly dependent on the leader. Such results suggest that benevolent leadership can regulate the function of authoritarian leadership on employee ethical behaviors by establishing certain disciplines and behavioral norms. Following this logic, when an authoritarian leader exhibits low benevolence, subordinates would understand that not completing tasks or not conforming to the leader's orders will definitely invoke punishments. Accordingly, they would have strong incentives to displace the responsibility of their behavioral choices to the leader. In contrast, when the authoritarian leader exhibits high benevolence and builds strong guanxi with employees, employees may feel strong obligations to the leader and thus are less likely to displace the responsibility of their behavior to the leader (Takeuchi et al., Reference Takeuchi, Wang and Farh2020). Instead, they would look for other ways to improve their performance (i.e., taking more initiatives in self-learning and making innovations at work).
Last, we examined the relationship between authoritarian leadership and UPB based on the sale context in China. Due to intensifying competition, both leaders and employees are motivated to improve their sales performance. Thus, leaders may become authoritarian and employees may engage in UPB. This possibility may result in the overestimation of this relationship in Study 2. Meanwhile, compared to their counterparts in lower power distance societies, Chinese employees accept more hierarchical differences and tolerate the command-control leadership style (Hofstede & Minkov, Reference Hofstede and Minkov1991). Thus, they may be more likely to attribute the responsibility of their unethical behaviors to the authoritarian leader. Considering such possibilities, we encourage researchers to examine the focal relationships we investigated and examine their generalizability in other contexts.
CONCLUSION
Regarding the mixed evidence on the effect of authoritarian leadership, our research demonstrates that authoritarian leadership motivates individuals to displace responsibility for their actions to the leader, thereby promoting UPB at work. Meanwhile, those who have a strong sense of moral efficacy tend to display a high level of self-directedness and are less likely to be influenced by an authoritarian leader. Our results thus extend the knowledge of authoritarian leadership by highlighting ‘harmful compliance’ and provide useful insights for understanding the effects of authoritarian leadership in Chinese societies. Managers should be aware of this possibility and take steps to avoid its negative influences on subordinates’ unethical behaviors.
APPENDIX I
Experimental Scenarios for Study 1*
High authoritarian leadership
YK is a software development company dedicated to providing management system and mobile phone software customization service for clients. The company has roughly 200 employees. You are a salesperson in Department of Marketing. It is the second year since you started to work in this company. Over the last two years, you have found that Liu Yang, the manager of Department of Marketing Development, always imposed paternalistic control at work. All matters in the department are decided on his own. He never shares information with subordinates in the department. In meetings, in case of different viewpoints, everyone is required to take Liu Yang's ideas as the final decision. Further, Liu Yang always tends to be rather strict, requiring the subordinates to adhere to the high sales goals he made. In case that staff fail to complete sales tasks, Liu Yang will punish them in a rather severe way without considering their feelings. Those who work with him have to handle tremendous pressure.
Low authoritarian leadership
YK is a software development company dedicated to providing management system and mobile phone software customization service for clients. The company has roughly 200 employees. You are a salesperson in Department of Marketing. It is the second year since you started to work in this company. Over the past two years, you have found that Liu Yang, the manager of Department of Marketing, is quite democratic in making decisions at work. All matters in the department are discussed and decided by group members. Liu Yang always shares important information with subordinates in the department. In meetings, he keeps an open mind and listens to different opinions of subordinates. Further, he is always kind and his subordinates feel free to propose their own suggestions on his sales goals. If sales tasks are not accomplished, Liu Yang will not scold his subordinates. Instead, Liu Yang will try to learn more about specific difficulties and problems for them. For this reason, the subordinates in the departments all feel quite relaxed working with him.
*Notes: To eliminate the possible gender effect, Liu Yang was referred to as male in the scenario for a half of the sample, and as female for the other half of the sample.
