More humility, less counterproductive work behaviors? The role of interpersonal justice and trust

Abstract

Leader humility has emerged as an important topic in understanding the role of leadership in organizations in recent years. Though it was found to enhance subordinates’ work performance and positive work behaviors, we are unaware of the psychological mechanism and boundary conditions underlying leader humility and employees’ negative behaviors toward leaders. Drawing on social exchange theory and using a multistage sample including 273 employees and 55 supervisors in China, we demonstrated a negative indirect effect between leader humility and subordinates’ counterproductive work behaviors toward supervisors (CWB-S) via interpersonal justice and trust in supervisor. Furthermore, we find that leader political skill moderates the effect of leader humility on interpersonal justice and trust in supervisor. The indirect effect of leader humility on subordinates’ CWB-S through interpersonal justice is stronger when leader political skill is high, suggesting a moderated mediation model. Finally, we discuss the theoretical contributions and practical implications of this study, and highlight future directions for research on leader humility.

Introduction

In today’s business environment, which is characterized by increasing dynamics and uncertainty, it is becoming increasingly difficult for a leader to “figure it all out at the top” (Morris et al. 2005; Senge 1990, p. 7). While traditional leadership research emphasized the greatness of leaders, emerging attention is being paid to “leading from the ground” and the expressed humility of leaders (Owens and Hekman 2012). Expressed humility implies “(a) a manifested willingness to view oneself accurately, (b) a displayed appreciation of others’ strengths and contributions, and (c) teachability” (Owens et al. 2013, p. 1518). Previous studies demonstrated that expressed humility could lead to beneficial outcomes at the organizational level (such as top management team integration and empowering organizational climate, e.g., Ou et al. 2014), team level (such as team performance, e.g., Owens and Hekman 2016; Rego et al. 2017), and individual level (such as employee creativity, e.g., Wang et al. 2017). Owens et al. (2013) indicate that leader humility substantially shapes interpersonal interactions between leaders and followers, and thus affects perceptions, attitudes, and behavioral responses.

Accordingly, most current research on leader humility focuses on how it promotes followers’ positive behaviors toward their work and organizations, such as task performance (Diao et al. 2019; Yu and Wang 2017), voice (Li et al. 2019), helping behavior (Mao et al. 2017), and creativity (Lei et al. 2015). However, whether and how leader humility affects followers’ negative behaviors toward their supervisors remains unclear. It is vital to address this research question because employees’ negative behaviors might psychologically harm other members in the workplace and cause serious economic loss to organizations (Aquino and Douglas 2003; Robinson and O’Leary-Kelly 1998). In addition, examining employees’ direct reactions toward their leaders can deepen our understanding of the relationship-based feature of leader humility (Owens et al. 2013). Therefore, the present study intends to examine the influence of leader humility on followers’ counterproductive workplace behaviors toward supervisors (CWB-S), a type of negative behavior directed at a supervisor.

CWB-S refers to organizational members’ voluntary, potentially destructive or detrimental acts that harm their leaders (Spector and Fox 2002). As suggested in previous studies, the poor quality of social exchange is one major antecedent of CWB-S (Thau et al. 2009). In contrast, leader humility is more likely to engender supportive leader-follower relationships and thus reduces CWB-S (Morris et al. 2005; Peters et al. 2011). Although the extant research has introduced leader-member exchange (LMX) to explain the mechanisms underlying leader humility and employee work behaviors (Wang and Zhang 2018; Yu and Wang 2017), it has not elaborated the social exchange process between leader humility and employees’ supervisor-directed behaviors. According to de Coninck (2010), fairness and trust are two important facets that reflect the quality of social exchange. When employees perceive that they are fairly treated and view their leaders as trustworthy, they reciprocate by engaging in more positive behaviors and reducing negative behaviors (Dirks and Ferrin 2002; Sousa-Lima et al. 2013). Thus, the present study introduces interpersonal justice and trust in supervisor as two mediators to enhance understanding of the social exchange mechanisms underlying the relationship between leader humility and follower CWB-S.

Moreover, note that the consequences of leader humility might also be contingent. Leader humility is effective only when followers perceive the sincerity of the leader (Owens and Hekman 2012). One potential reason for effective humility might be leaders’ high political skill. Leader political skill refers to “the ability to effectively understand others at work, and to use such knowledge to influence others to act in ways that enhance one’s personal and/or organizational objectives” (Ahearn et al. 2004, p. 311). Higher political skill could show leaders’ apparent sincerity, social astuteness, interpersonal influence, and networking ability, which may further enhance the effectiveness of their humility by improving the credibility of their behaviors and reducing employees’ sense of being manipulated (Ferris et al. 2005, 2007).

To summarize, our proposed model is illustrated in Fig. 1. This study may contribute to the leader humility literature in the following ways. First, our research broadens the outcomes of leader expressed humility by demonstrating its role in decreasing CWB-S. We thereby provide one additional perspective to investigate the effect of leader humility on employees’ negative behaviors toward their immediate leaders. Second, our research unpacks underlying social exchange mechanisms by identifying two interpersonally oriented factors (i.e., interpersonal justice and trust in supervisor), rather than relying only on LMX. Therefore, our work enables specifically accounting for the relationship between leader humility and follower CWB-S from a social exchange approach. Third, this study further extends the boundary conditions for the effects of leader humility on employee behaviors. The buffering effect of leader political skill suggested in our study highlights the need for further studies to identify when leader humility is more effective.

Fig. 1

Theoretical model

Theory and hypotheses

Leader humility and CWB-S

As social exchange theory postulates, “reciprocity” is a crucial rule of social exchange. One person’s reactions resulting from the social exchange process are contingent on the actions of another person (Blau 1964). When leaders act kindly, their subordinates are likely to feel obligated to reciprocate the goodwill they receive, which reduces the possibility of hurting these leaders (Cropanzano and Mitchell 2005).

The self-transcendence feature of leader humility enables leaders to see value and worth beyond themselves (Grenberg 2005). By admiring employees’ strengths and appreciating their contributions, leader expressed humility enhances employees’ perceptions of a leader’s respect, appreciation, and goodwill in the LMX process (Owens and Hekman 2012). As reciprocity, employees are less likely to intentionally adopt CWB-S which may hurt their leaders and reduce leadership effectiveness (Fox et al. 2001). Therefore, we hypothesized the following:

Hypothesis 1: Leader humility is negatively related to subordinates’ CWB-S.

The mediating role of interpersonal justice

Compared to other types of organizational justice (distributive, procedural, and informational justice), interpersonal justice is central in influencing employee behaviors (Holtz and Harold 2013). The essence of interpersonal justice indicates the degree of employees’ perception of being treated with respect and consideration (Greenberg 1990). Leader humility including openness to feedback, appreciation of others, and fewer self-focused characteristics is likely to elicit more consideration and respect (Ou et al. 2014). More important, fairness is a salient interpersonal characteristic of humility. Humble individuals treat others as equals (Chancellor and Lyubomirsky 2013), are less likely to take advantage of others (Lee and Ashton 2004), and demonstrate more fairness and restraint (Rowatt et al. 2006).

From the social exchange perspective, interpersonal justice, which reflects the quality of the exchange between employees and their supervisors, would be positively associated with leader humility (Owens et al. 2013). Previous studies prove that interpersonal justice is negatively associated with CWB-S (Jones 2009). Therefore, when leaders express more humility, employees feel respected and cared about by leaders. Feeling respected further leads to a high perception of interpersonal justice (Wu et al. 2012). Accordingly, based on the principle of reciprocity, employees are less likely to engage in CWB-S when their leaders demonstrate higher humility (Eisenberger et al. 2004; Litzky et al. 2006; Thau et al. 2007). Thus, we assume that:

Hypothesis 2: Interpersonal justice mediates the negative relationship between leader humility and subordinates’ CWB-S.

The mediating role of trust in supervisor

Trust in supervisor is a relational construct that reflects employees’ evaluations of leaders’ trustworthiness based on their interactions with the supervisor (Luo 2005). The level of trust in supervisor may be strengthened or weakened by the experiences, interactions, and context in which the leader-member relationship exists (Burke et al. 2007). Leaders who express humility tend to develop objective assessments of themselves and disclose their personal weaknesses (Sprecher et al. 2013), often leading to reciprocal disclosure and increases in trust from employees (Collins and Miller 1994; Ehrlich and Graeven 1971). In addition, by adopting a less self-interested leadership approach and showing teachability, leader humility can reduce followers’ sense of uncertainty in understanding leaders’ behaviors and foster followers’ trust in supervisor (Kenworthy and Jones 2009; Nielsen et al. 2010; Owens et al. 2013).

Trust in supervisor reflects the quality of the social exchange between leaders and followers. When followers trust their leaders, they are more aware of how to develop and maintain the LMX relationship such as engaging in more positive behaviors (such as organizational citizenship behavior) and reducing negative behaviors (such as CWB-S, interpersonal deviance; Dirks and Ferrin 2002). Therefore, according to social exchange theory, when leaders express more humility, employees are motivated to trust their leaders and reciprocate by engaging in beneficial rather than negative behaviors (Zapata et al. 2013). As such, we propose the following hypothesis:

Hypothesis 3: Trust in supervisor mediates the relationship between leader humility and subordinates’ CWB-S.

The moderating effect of leader political skill

Previous research on social influence demonstrate that political behaviors might enhance the perceptions of justice in organizations (Gavin et al. 1995). Leaders with high political skill tend to have social acuity and interpersonal astuteness (Ferris et al. 2005). They are more competent in adopting the most appropriate ways to express humility according to different situational characteristics, which can make most employees feel comfortable, perceive interpersonal respect, and develop good rapport with leaders (Ahearn et al. 2004; Ewen et al. 2013). In addition, humble leaders with high political skill are more able to enhance employees’ perceptions of sincerity rather than ingratiation (Ahearn et al. 2004), and strengthen their feeling of concern, respect, and fairness from the leader (Treadway et al. 2004). Thus, the positive effects of leader humility on perceptions of interpersonal justice can be strengthened by a leader’s high political skill.

In contrast, leaders with low political skill are less effective in making employees feel sincere. Leaders who disguise praise to win favor and create a “humble” image are less likely to be truly appreciated by employees, or may push employees to be defensive and cautious (Ahearn et al. 2004). Thus, when leaders with low political skill express humility, employees feel less cared about and more controlled. This then attenuates the positive effects of leader humility on perceptions of interpersonal justice. Thus, we hypothesized that:

Hypothesis 4: Leader political skill positively moderates the positive relationship between leader humility and interpersonal justice, such that the positive relationship is stronger when the leader has a higher level of political skill.

Leaders with a high level of political skill are good at getting their employees to respond positively (Ahearn et al. 2004). When leaders possess a high level of political skill, employees perceive more respect and fair treatment, and less manipulation or controlling behavior (Ferris et al. 2005). Therefore, the mediating effect of interpersonal justice between leader humility and subordinates’ CWB-S is stronger when leaders have a high level of political skill. Conversely, leaders with a low level of political skill could make employees perceive lower sincerity and respect than those with high political skill, and the mediating role of interpersonal justice between leader humility and employee CWB-S would be lower. Therefore, we hypothesize that:

Hypothesis 5: Leader political skill moderates the indirect effect of interpersonal justice between leader humility and followers’ CWB-S, such that the indirect effect is stronger when leader political skill is high and weaker when it is low.

Leaders with a high level of political skill not only know what to do in different situations at work, but also understand how to act in a sincere manner that inspires believability, trust, and confidence (Ahearn et al. 2004). When politically skilled leaders express humility, their self-disclosure to and appreciation of others are more likely to be viewed as genuine and sincere. Such actions reflect leaders’ influence attempts, rather than an individual exhibit, without an ulterior motive (Ferris et al. 2005). Thus, humble leaders with a high level of political skill are more prone to obtain employees’ trust (Smith et al. 2009). In addition, political skill gives people a sense of self-confidence and personal security (Ahearn et al. 2004). When the competence and personal security of humble leaders are perceived, employees are willing to trust their leaders with fewer concerns about potential risks (Owens and Hekman 2012).

Compared to leaders with a high level of political skill, those with a low level are less effective in understanding others and conveying a sense of personal security (Ferris et al. 2005, 2007). Humble leaders with low political skill are more likely to be viewed as possessing some ulterior self-serving motive than those with high political skill (Ferris et al. 2005), which may make employees feel at risk and reduce their trust in leaders. As such, we propose the following hypothesis:

Hypothesis 6: Leader political skill positively moderates the relationship between leader humility and trust in supervisor, such that the positive relationship is stronger when the leader has a higher level of political skill.

When leaders with a high level of political skill express humility, employees believe that leaders’ focus is outward, on others rather than themselves; thus, they have more trust in their leaders (Ahearn et al. 2004; Ferris et al. 2005). Since trust is considered the basis of followers’ work intentions and behaviors (Litzky et al. 2006), employees’ likelihood to engage in CWB-S decreases when supervised by leaders considered humble with a higher level of political skill. However, leaders’ low political skill may reduce employees’ feeling of interpersonal security and increase their perceptions of uncertainty regarding their leaders’ behaviors. Thus, a low level of political skill can decrease the indirect effect of leader humility on subordinates’ CWB-S via trust in supervisor. Thus, we propose the following hypothesis:

Hypothesis 7: Leader political skill moderates the indirect effect of trust between leader humility and followers’ CWB-S, such that the indirect effect is stronger when leader political skill is high and weaker when it is low.

Method

Sample and procedure

We tested the hypotheses using data collected from two Chinese manufacturing companies in (in Beijing and Luoyang, respectively). Participants in our study were all full-time employees and their immediate leaders who engage in various functions including (a) research and development (R& D), (b) marketing and sales, (c) engineering or project operation, and (d) administration. In both companies, with the support of the CEOs, the human resource departments provided the name lists of participants to our research team and scheduled times to deliver the questionnaires. On the day of data collection, all participants received questionnaires with envelopes from our research team, responded to their own questionnaires voluntarily and faithfully, and returned the surveys directly to our research team in sealed envelopes after completing them. Our research team assured participants of the confidentiality of their responses. To minimize potential common method variance and test the causality among variables, we used two sets of questionnaires (one for employees and the other for their immediate supervisors) and matched them using prepared codes (Podsakoff et al. 2003). We collected our data in two waves. At time 1, employees provided information on their demographics and leader humility, while supervisors reported their demographics and political skill. At Time 2, employees reported their perceptions of interpersonal justice and trust in supervisor, while their respective supervisors evaluated subordinates’ CWB-S.

Data were collected from 320 employees in 66 teams, and 284 employees from 65 teams returned the questionnaires after the two phases, resulting in a response rate of 88.8% at the individual level and 98.5% at the team level. After eliminating incomplete and unmatched questionnaires, we obtained valid responses from 273 employees and 55 supervisors (response rates of 85.3% and 83.3%, respectively). Of the 273 employees, 55.7% were male and the average age was 35.57 years old (SD = 7.57). Furthermore, 63.2% held an under-bachelor’s degree, 32.4% a bachelor’s degree, and 4.4% a master’s degree. Of the 55 supervisors, 65.5% were male and the average age was 40.04 years old (SD = 6.73). In addition, 60.0% held an under-bachelor’s degree, 34.5% a bachelor’s degree, and 5.5% a master’s degree. Each team had only one official leader and the leaders had very close and frequent interactions with their subordinates at work.

Measures

Because the survey was originally developed in English, we followed a standard translation and back-translation procedure (Brislin 1986) to develop Chinese measures. All items were assessed on a six-point Likert scale ranging from 1 (strongly disagree) to 6 (strongly agree).

Leader humility

We used a nine-item scale from Owens et al. (2013) to assess employees’ perceptions of leader humility. A sample item was “My supervisor actively seeks feedback, even if it is critical.” Cronbach’s α for this scale was 0.92.

Interpersonal justice

Interpersonal justice was measured on a four-item scale developed by Colquitt (2001). A sample item was “My leader treated me in a polite manner.” Cronbach’s α for this scale was 0.89.

Trust in supervisor

We measured trust in supervisor using McAllister’s (1995) eight-item scale. Sample items included “We have a sharing relationship. We can both freely share our ideas, feelings, and hopes” and “My leader approaches his/her job with professionalism and dedication.” Cronbach’s α for the scale was 0.93.

CWB-S

We used Jones’ (2009) five-item measure of CWB-S that referred to an immediate supervisor. A sample item was “Purposely neglected to follow your instructions.” Cronbach’s α for this scale was 0.86.

Leader political skill

We assessed leader political skill on a six-item scale developed by Ahearn et al. (2004). A sample item was “I am good at getting others to respond positively to me.” Cronbach’s α for this scale was 0.75.

Control variables

Following previous studies, we controlled employees’ and supervisors’ age, gender, and education level in our analyses, because these variables were found to directly influence subordinates’ CWB (Jones 2009; Zhang and Deng 2016). Moreover, as we collected data from two companies, a dummy variable (1 = the manufacturing company in Beijing; 0 = the manufacturing company in Luoyang) was used to control the differences between the two companies.

Results

Confirmatory factor analysis

Multilevel Confirmatory Factor Analyses (MCFAs) were conducted to examine the hypothesized five-factor structure while accounting for the nested structure of the data (Dyer et al. 2005; Muthén 1994).

The fit indices reveal that the proposed five-factor model fits the data well: χ² ₍₁₂₈₎ = 226.98, p < 0.001, CFI = 0.95, TLI = 0.94, RMSEA = 0.05, SRMR_within = 0.05, and SRMR_between = 0.06. Moreover, the baseline model yields the best fit when compared with a four-factor model in which trust in supervisors and CWB-S were set to one latent variable (χ² ₍₁₃₅₎ = 304.67, p < 0.001, CFI = 0.92, TLI = 0.90, RMSEA = 0.07, SRMR_within = 0.08, SRMR_between = 0.27); a three-factor model in which interpersonal justice, trust in supervisors, and CWB-S were set to one latent variable (χ² ₍₁₄₀₎ = 427.02, p < 0.001, CFI = 0.87, TLI = 0.84, RMSEA = 0.09, SRMR_within = 0.09, SRMR_between = 0.29); and a two-factor model in which all individual-level variables were set to one latent variable (χ² ₍₁₄₃₎ = 711.89, p < 0.001, CFI = 0.74, TLI = 0.69, RMSEA = 0.12, SRMR_within = 0.11, SRMR_between = 0.32) (Table 1).

Table 1 Multilevel confirmatory factor analysis

Descriptive statistics

Table 2 provides the means, standard deviations, reliabilities, and correlations. As Table 2 shows, leader humility is positively related to interpersonal justice (r = 0.52, p < 0.01) and trust in supervisor (r = 0.63, p < 0.01). Employees’ CWB-S is negatively related to interpersonal justice (r = − 0.21, p < 0. 01) and trust in supervisor (r = − 0.14, p < 0. 05), which provides initial support for our proposed relationships.

Table 2 Means, standard deviations, reliabilities, and correlations among study variables

Hypothesis testing

Given our nested data structure, we used hierarchical linear modeling (HLM) analyses with HLM 6.08 software to test our hypotheses (Raudenbush et al. 2004). The test results of the null models suggest that the between-group variance in subordinates’ CWB-S (χ² ₍₅₄₎ = 1166.59, p < 0.01, ICC (1) = 0.83), interpersonal justice (χ² (54) = 81.11, p < 0.05, ICC (1) = 0.10), and trust in supervisor (χ² ₍₅₄₎ = 90.45, p < 0.01, ICC (1) = 0.12) was significant, supporting the use of HLM in this study.

Hypothesis 1 proposes that leader humility is negatively related to followers’ CWB-S. The results of model 2 show that leader humility is significantly related to CWB-S (γ = − 0.05, p < 0.05). Thus, Hypothesis 1 is supported.

Hypotheses 2 and 3 predict that interpersonal justice and trust in supervisor both mediate the relationship between leader humility and subordinates’ CWB-S. Following Baron and Kenny’s (1986) analytic strategy for testing mediating effects, in Table 3 we first find that leader humility is significantly and positively related to interpersonal justice (Model 6, γ = 0.49, p < 0.01) and trust in supervisor (Model 8, γ = 0.67, p < 0.01). When both leader humility and interpersonal justice are included in the regression model, interpersonal justice is significantly and negatively related to CWB-S (Model 3, γ = − 0.11, p < 0.05), while the effect of leader humility on CWB-S becomes insignificant (Model 3, γ = − 0.01, n.s.). When both leader humility and trust in supervisor are included in the regression analyses, trust in supervisor is significantly and negatively related to CWB-S (Model 4, γ = − 0.08, p < 0.05), whereas the effect of leader humility on CWB-S becomes insignificant (Model 4, γ = − 0.01, n.s.). These findings indicate that interpersonal justice/trust in supervisor mediates the relationship between leader humility and CWB-S. We also conduct Monte Carlo tests with 20,000 times bootstrapping (Selig and Preacher 2008) to further examine the mediation effects. The results show that the indirect effects of leader humility on CWB-S via interpersonal justice (95% CI = [− 0.073, − 0.009], not containing zero) and via trust in supervisor (95% CI= [− 0.074, − 0.002], not containing zero) are significant. Thus, hypotheses 2 and 3 are supported (Table 3).

Table 3 Results from HLM analysis—Mediating effects

Hypotheses 4 and 6 propose that leader political skill amplifies the relationship between leader humility and interpersonal justice/trust in supervisor. The results of Models 10 and 12 in Table 4 show that the interaction of leader humility and leader political skill is significantly and positively related to interpersonal justice (γ = 0.20, p < 0.01) and trust in supervisor (γ = 0.26, p < 0.01). We then conduct simple slope tests and plot the positive moderating effects of leader political skill following the suggestions of Aiken and West (1991). Figs. 2 and 3 show that the relationship between leader humility and interpersonal justice is stronger when leader political skill is high (γ = 0.70, p < 0.01) than when it is low (γ = 0.29, p < 0.01). The relationship between leader humility and trust in supervisor is stronger when leader political skill is high (γ = 0.93, p < 0.01) than when it is low (γ = 0.41, p < 0.01). Therefore, hypotheses 4 and 6 are supported.

Table 4 Results from HLM analysis—Moderating effects

Fig. 2

Interactive effect of leader humility and leader political skill on interpersonal justice

Fig. 3

Interactive effect of leader humility and leader political skill on trust in supervisor

Hypotheses 5 and 7 state that leader political skill moderates the indirect effect of leader humility on subordinates’ CWB-S via interpersonal justice/trust in supervisor. We test these two hypotheses according to the moderated mediation model (Edwards and Lambert 2007).

As shown in Table 5, the indirect effect of leader humility on CWB-S via interpersonal justice is significant when leader political skill is both high (γ = − 0.15, p < 0.05, 95%CI = [− 0.25, − 0.04]) and low (γ = − 0.13, p < 0.05, 95%CI = [− 0.21, − 0.04]). In addition, the difference in the indirect effects of leader humility is marginally significant (∆γ = − 0.02, p < 0.10, 95%CI = [− 0.04, − 0.01]). Thus, hypothesis 5 is supported.

Table 5 Results of the moderated path analysis

Table 6 shows that the indirect effect of leader humility on CWB-S via trust in supervisor is marginally significant when leader political skill is both high (γ = − 0.13, p < 0.10, 95%CI = [− 0.26, − 0.01]) and low (γ = − 0.12, p < 0.10, 95%CI = [− 0.22, − 0.01]). However, the difference in the indirect effects of leader humility is not significant (∆γ = − 0.02, n.s., 95%CI = [− 0.04, 0.01]). Thus, hypothesis 7 is not supported.

Table 6 Results of the moderated path analysis

Discussion

Drawing on social exchange theory and using a sample of 273 supervisor-subordinate dyads in China, we propose and test a model on how and when leader expressed humility decreases subordinates’ CWB-S. We find that (a) leader humility is negatively related to subordinates’ CWB-S and (b) both interpersonal justice and trust in supervisor mediate this relationship separately. Furthermore, (c) leader political skill positively moderates the relationships between leader humility and interpersonal justice, and between leader humility and trust in supervisor such that positive relationships are stronger when leader political skill is high than when it is low. Finally, we find that (d) leader political skill moderates the indirect relationship between leader humility and CWB-S via interpersonal justice, such that the indirect effect of leader humility is stronger under conditions of high leader political skill. However, the moderated mediation effect of trust in supervisor is not significant. The findings of this research generate several valuable theoretical and practical implications.

Theoretical implications

Our study contributes to the existing literature in three significant ways. First, this study enhances our understanding of the role of leader humility by broadening the range of its individual consequences. Existing studies establish various positive impacts of leader humility on employees’ performance and positive behaviors toward their work (e.g., creativity, Wang et al. 2017; proactive behavior, Chen et al. 2018; voice, Bharanitharan et al. 2018). Nonetheless, little is known about the potential impact of leader humility on employees’ behaviors toward their leaders. Based on social exchange theory, our results suggest that leader expressed humility can effectively reduce employees’ intentional destructive behaviors toward their leaders. This study provides empirical evidence to support the notion that leader humility does not only promote employees’ positive attitudes toward their work and performance, as shown in past studies, but also inhibit their negative behaviors toward their leaders. More important, because employees’CWB is a common event in today’s workplaces, there is growing interest in studying the antecedents of CWB (Harold et al. 2016; Yang and Treadway 2018). This study also contributes to research on CWB by explicitly differentiating directly affective targets (organization or leader, Jones 2009) and exploring the antecedents of CWB-S.

Second, building on social exchange theory, we elucidate the critical psychological mechanisms linking leader humility and CWB-S. In this study, we focus on the expressed, interpersonal nature of humility rather than intrapersonal, internal, and personal traits (Owens et al. 2013). Therefore, it is important to explore the social exchange process between leader humility and subordinates’ CWB-S. Previous studies propose that high social exchange is characterized by many aspects including care, trust, respect, and fairness (de Coninck 2010; Dirks and Skarlicki 2004). However, it is far from adequate to only introduce a generalized relationship quality concept like LMX to examine the influence of leader expressed humility on employee behaviors. The current study introduces interpersonal justice and trust in supervisor as two mediators to identify the specific influencing process of leader humility on subordinates’ CWB-S. In addition, the results of our moderated mediation tests show that the mediating effect of trust in supervisor is more stable than that of interpersonal justice, because it is not significantly influenced by a leader’s level of political skill. To this end, our study further clarifies and demonstrate the difference between justice-based and trust-based mechanisms.

Third, we contribute theoretically by identifying the boundary condition of the effect of leader humility. As organizations are considered political arenas (Mintzberg 1985), a lack of interpersonal or social skill may “derail” managers’ careers (van Velsor and Leslie 1995). Thus, Ahearn et al. (2004) call for further research to identify the processes between leader political skill and important employee reactions. This study addresses this appeal by examining how the intensity of the relationship between leader humility and interpersonal justice/trust in supervisor differs between leaders with different levels of political skill. We propose and demonstrate the moderating role of leader political skill in shaping followers’ perceptions and behaviors toward leaders who express humility.

Practical implications

Our study highlights significant practical implications for organizations, managers, and employees.

For organizations, CWB is common in many work teams, which might have financial costs and result in personal impacts on productivity and employees’ well-being (Penney and Spector 2005). It is necessary to find effective ways to reduce employees’ CWB toward organizations and leaders. Our results imply that leader expressed humility can effectively reduce followers’ CWB-S, suggesting that organizations, especially human resource departments, should strengthen the evaluation of leader expressed humility in selecting and developing potential leaders. In addition, behavioral patterns of leader humility can be cultivated through training (Owens et al. 2013). Therefore, organizations should increase their investment in the training process to teach managers how to regard themselves, others, and the environment.

For managers, this study finds that leader humility effectively reduces employees’ CWB-S by improving employees’ interpersonal justice and trust in supervisor. According to Owens et al. (2013), leader humility should be externally observable by others. Therefore, our study should serve as a reminder to managers to express more humble behaviors. In addition, leaders with high political skill can have a higher interpersonal reputation and organizational power, and elicit better team performance (Ahearn et al. 2004; Treadway et al. 2004). Therefore, supervisors should consider the influence of leaders’ political skill on the relationship between leadership style and employee behaviors toward both work and supervisors, and develop the ability to effectively persuade, influence, and control others (Mintzberg 1985).

For employees, our study suggests a path by which they can reduce their CWB-S when perceiving leader humility. Employees should be aware of the adverse effects of CWB-S and build high-quality interpersonal relationships with their leaders by following the rule of reciprocity in the social exchange process (Blau 1964).

Limitations and future research directions

Our study has several noteworthy limitations. First, although we draw on social exchange theory and introduce two mediators—interpersonal justice and trust in supervisor—to explore the underlying process of leader humility and subordinates’ CWB-S, future studies should consider alternative theories and mediators. For instance, based on justice theory, future studies could explore the relationship between leader humility and CWB-S via other types of justice such as distributive, procedural, and informational justice (Colquitt 2001). Furthermore, since our research only introduces mediating mechanisms at the individual level, we encourage future studies to explore other mediators at the team level. For example, leader humility may reduce team members’ CWB-S by increasing team humility (Rego et al. 2017) and team psychological safety (Hu et al. 2018).

Second, we only introduce CWB-S as one outcome. Thus, one potential direction for future study is to identify other types of followers’ behaviors toward supervisors, such as incivility (Andersson and Pearson 1999), bullying (Einarsen et al. 2009), and gossip (Kurland and Pelled 2000).

Third, in the current study, supervisors rate their own political skill, which may differ from that perceived by employees. Compared to leaders’ own evaluations of their behaviors, the perceptions of employees might have a more direct impact on their behaviors. Thus, we encourage scholars to examine the moderating effect of employee-rated leader political skill and compare these results with those of this study. Moreover, employees’ own political skill is also a key factor influencing the effectiveness of leader humility. Employees with a high level of political skill can understand leaders effectively and are more likely to behave appropriately in a given context (Ferris et al. 2005). We appeal to future researchers to further explore the role of employees’ political skill in the relationship between leader humility and employees’ CWB-S, as well as other employee behaviors toward leaders.

Finally, it is worth considering the influences of cultural and industrial characteristics in leader humility research. Previous studies confirm that perceptions of humility differ between individualist (e.g., the United States) and collectivist cultures (e.g., China) (Ou et al. 2014). Future studies are encouraged to test whether the findings of our study can be replicated in other cultures or different industries (power centralization vs. power decentralization).

Conclusion

Drawing on social exchange theory, this study proposes a psychological process model to explain how and when leader humility affects employees’ CWB-S. The current study introduces two variables in the leader-member interaction process, namely interpersonal justice and trust in supervisor to specifically unpack the psychological mechanism underlying leader humility and CWB-S. Finally, this study examines the moderating effect of leader political skill in a complex environment. This study finds that leader humility can effectively decrease subordinates’ CWB-S through increasing employees’ interpersonal justice and trust in supervisor, and the indirect effect of interpersonal justice is stronger when leaders have a high rather than low level of political skill. We call on future research to continue to explore the effect of leader humility on followers’ behaviors toward supervisors, considering the abundance of research opportunities in this area.