I Treated the Way You Treated Me: The Effect of Leader Hypocrisy on Employees’ Voice Behavior

Introduction

The contemporary business circumstance with uncertainty, complexity, ambiguity, and mutability pushes organizations to adapt to the rapidly changing environment. These changes also present a considerable challenge to managers, as demands on them to collect, process, and consolidate information are more daunting than ever. In such a situation, employees’ voice behavior, which refers to proactive behavior consenting to express thoughts and suggestions for upgrading the current organizational status,^1,2 becomes increasingly important to managers.³ By catching the ideas and information provided by the voicers, managers are able to adjust their actions and come up with more effective improvements than before.⁴ Besides benefiting managers and organizations, voice has many advantages. For individuals, such benefits include enhancing employees’ work engagement,⁵ job satisfaction,⁶ creativity,⁷ and work performance;⁸ from an organizational perspective, voice behavior can increase the efficiency and accuracy of organizational decision,⁹ organizations’ creativity performance,¹⁰ organizational teams’ innovation level and team performance.¹¹

Being aware of the numerous benefits brought by voice behavior, researchers have put insight into antecedents of voice behavior from multiple perspectives, such as individual characteristics (eg, proactive personality,¹² cognition and affection,^13,14 capability variance¹⁵), and organizational context and design.^16,17 As leaders play an essential role as receivers and executors of the voice, employees speak more or less largely depending on leadership style and leaders’ characteristics.^1,4 However, despite that the decisive role of leaders in employee voice behavior has also been extensively confirmed, previous studies have largely concentrated on the impact of leadership (which captures comprehensive characteristics of leaders, such as humble leadership,¹⁸ ethical leadership,¹⁹ and servant leadership²⁰) but given insufficient consideration to the unique and specific behavioral traits of leaders.

Considering that the primary goal of voice is to change current states,²¹ one of the main concerns held by the employees is whether leaders can effectively implement their suggestions and address the questions.²² Therefore, in the voice process, employees’ judgments of a leader’s integrity or hypocrisy are important for that employees are usually unwilling to voice when leaders’ responses are uncertain.²³ This suggests that leader hypocrisy, which refers to an inconsistency between a leader’s words and behavior,²⁴ might serve as an “intangible closed door” that hampers employees’ willingness to speak up. However, current research fails to reveal the potential relationship between leader hypocrisy and employee voice behavior, as well as the influence mechanism and its’ boundary conditions. This underlines the importance of exploring the impact of leader hypocrisy on employees’ voice behavior, which may provide insightful perspectives to unveil the barriers that prevent leadership from having a downward impact on employees.

Therefore, we draw upon social cognitive theory to explore the influence of leader hypocrisy on voice behavior. As Bandura²⁵ stated, environmental factors will eventually trigger changes in their behavior by influencing the inner state of the individuals. According to the theory, we assume leaders’ word-deed inconsistency might damage employees’ cognition-based and affect-based trust in their leaders. Cognition-based trust comes from the individual’s evaluation of the ability and reliability of others, while affect-based trust comes from the social interaction between the individual and others.^26,27 Following this, on the one hand, employees evaluate those leaders who do not perform their words with low credibility, which reduces employees’ evaluation of the credibility of leaders as well as damages employees’ cognition-based trust.²⁸ Meanwhile, hypocritical behavior, to a large degree, violates the foundation of a well-built interpersonal relationship between leaders and employees, which reduces employees’ emotional dependence on leaders, and the changes will then be reflected in the behavior of employees. When employees have low cognition-based and affect-based trust toward their leaders, they are unwilling to take risks getting along with leaders²⁹ and are less likely to implement risky voice behaviors. Therefore, our study argues that cognition-based trust and affect-based trust mediate the relationship between leader hypocrisy and voice behavior.

Moreover, social cognitive theory indicates that individuals with different characteristics evoke different psychological reactions toward the environment.³⁰ Although the inconsistency between words and behavior cannot reflect personal moral attitudes, the inconsistent behavior violates the moral norms related to honesty and integrity,³¹ which makes it possible that individuals’ thoughts on the inconsistent behavior will directly affect their views on this behavior. Therefore, we argue that moral identity will affect the strength of the impact of leadership hypocrisy on employees’ cognition-based trust and affect-based trust. Moral identity is an individual’s self-concept on a series of moral traits such as honesty and fairness, reflecting the importance of those moral qualities to an individual’s self-concept.³² The higher the individual’s moral identity, the heavier the weight of moral traits on an individual.³² Following this, our study argues that compared with employees with low moral identity, employees with higher moral identity have lower cognition-based trust and affect-based trust in hypocritical leaders. Therefore, this study reveals the impact of leadership hypocrisy on employee voice behavior through cognition-based trust and affect-based trust and further introduces the moderating effect of employees’ moral identity. The theoretical model is shown in Figure 1.

Figure 1 Theoretical model.

Our research adds to the existing literature on voice behavior in various ways. First, we extend the research on the interaction between leaders’ behaviors and followers’ voices by illustrating a unique phenomenon of hypocrisy. Although previous research extensively tested the impact of leader-related factors on voice behavior,^33–35 there still exists a need for more in-depth and refined examinations of the antecedents of voice behavior by delineating the behavior of leaders from a micro perspective. Building upon the close relationship between leaders’ reactions and the employee’s voice behavior, we extend voice research by examining the effect of leader hypocrisy (a pervasive, yet less noticed, aspect of leadership behavior) on employees’ voice behavior. In doing so, our research not only adds to existing research on the antecedents of employees’ voice behavior but also sheds light on the potential damage brought by the leader’s “fake behavior”, thereby responding to the calling of exploring the consequences of lousy leader behavior,³⁶ and revealing the diverse perceptions of employees facing leader’s behavior.³⁷

Secondly, our research reveals a dual pathway by which leader hypocrisy can prohibit proactive organizational behavior, which can provide researchers with a comprehensive understanding of the question, “How does leader hypocrisy work?”. Although the role of leadership hypocrisy in triggering negative employee behavior has been widely verified,^24,38,39 few studies have discussed the changes in employees’ internal state after experiencing leader hypocrisy, which has led to a poor understanding of why and how could leader hypocrisy trigger negative employee reactions. In this study, we explain the adverse effects of leadership hypocrisy in terms of cognitive and affective aspects, drawing upon social cognition theory. By interpreting the employees’ psychological changes when confronted with leader hypocrisy, our study provides scholars with a new mechanism to further examine the role of leader hypocrisy in the workplace.

Third, we extend the boundary condition on the effect of leader hypocrisy from followers’ personal trait aspect. Leader hypocrisy occurs in the interaction of leader and employees, and different employees have diverse interpretations. Based on the social cognition theory, we introduce moral identity into the research framework of leader hypocrisy. Our findings also confirm that followers with different levels of moral identity would respond differently to leader hypocrisy. By doing so, we explore the boundary conditions under which leadership hypocrisy operates from the perspective of followers’ traits.

Hypothesizes

Leadership Hypocrisy and Employee Trust

At the individual level, hypocrisy is considered to be a discrepancy between an individual’s propagandized views and his or her actual behavior, manifested in a series of inconsistent behaviors such as treating others strictly and being lenient with themselves and “saying one thing but behave differently”.⁴⁰ Such behavior occurs at a high frequency among leaders,²⁴ and Effron, O’Connor, Leroy, and Lucas⁴¹ suggested that the high frequency may be due to the unique job nature of the leader in the organization. As a communicator, leaders must interact with various organizational stakeholders and weigh the pros and cons from different views before deciding. The uncertainty could lead to final decisions violating their previous value propositions and promises.⁴² Although the concept of leadership hypocrisy itself cannot be defined as good or bad, hypocritical behavior is usually perceived as the reflection of unfavorable characteristics such as falsehood, cheating, and concealment.⁴³ Thus, the inconsistency between leaders’ behaviors and their intentions tends to trigger employees’ negative evaluations, which in turn causes a series of adverse effects such as employees’ negative emotions, turnover intention, reduction in organizational citizenship behavior, and increased loafing behavior.^24,38,44

Trust is defined as the degree to which an individual believes in the words, decisions, and behaviors of others and is willing to act accordingly.²⁶ In the process of interpersonal communication, the establishment of the trust relationship is often derived from an individual’s evaluation of the trustworthiness of each other, which is closely related to the characteristics of the trust object in daily behavior.²⁷ McAllister²⁷ pointed out two different forms of trust in organizations: cognition-based trust and affect-based trust. Cognition-based trust derives from a rational judgment toward evaluating one’s abilities, reliability, honesty, and integrity.²⁷ Moreover, the building of affect-based trust comes from the emotional resonance of individuals in terms of empathy, belonging, and respect, which can be seen as a stable and deep emotional bonding established by the two parties in the interaction process.²⁷ Scholars have widely recognized McAllister²⁷ and have applied this classification to the studies of superior-subordinate relationships in organizations.^45,46 Previous scholars found that cognition-based and affect-based trusts are critical factors in bridging the leaders’ and employees’ performance (eg, performance improvement,⁴⁷ organizational citizenship behavior.⁴⁸ However, different factors bring different influences on the two types of trust. Studies have shown that factors that reflect trustiness, such as transformational leadership, competence, and behavioral integrity, significantly enhance individuals’ cognition-based trust;^49,50 while those emotional-related factors, such as servant leadership, benevolence, and value alignment, positively affect individuals’ affect-b trust.⁵¹

Accordingly, we assume employees’ cognition-based trust in leaders will decline when faced with leadership hypocrisy. Leadership hypocrisy is categorized as deceitful behavior that contradicts the leader characteristics that employees expect, such as leadership integrity and honesty.³⁹ The inconsistency leads employees to label the leader’s behavior as a “hypocrite”. Under such a situation, employees will doubt the leader’s ability and dependability, reducing their cognition-based trust toward the leader. At the same time, leadership hypocrisy increases the unpredictability of leadership behavior. Predictability is one of the crucial bases for establishing a cognition-based trust relationship between two parties, and it is equally important as the reliability of the trustee.⁵² Since leadership hypocrisy represents the behavior inconsistency between the leader’s words and his or her behavior, employees would be unable to predict the leader’s future behavior.²⁴ In turn, the uncertainty increases employees’ doubt about the correctness and effectiveness of leaders’ decisions, ultimately leading to decreased cognition-based trust. Hence, our research hypothesis:

Hypothesis 1a: Leader hypocrisy negatively influences employees’ cognition-based trust.

On the other hand, we assume leader hypocrisy will damage the affect-based trust between leaders and employees. First, leaders’ negative behavior is a significant resource for employees’ negative emotions.⁵³ When perceiving the leader’s hypocritical behavior, we assume employees tend to regard it as a selfish expression that leaders tend to maximize their self-interests by violating the original value proposition and putting their interests above the benefits of employees. Such a situation could evoke negative emotions (eg, anger and disappointment) in employees, leading to decreased affective trust in the leader.⁵⁴ Second, reciprocity is one of the critical principles for establishing affective trust.⁵⁵ Indeed, the perception of being respected, trusted, and treated equally is an important signal for individuals to maintain affective trust with others.⁵⁶ When employees perceive their leaders’ behavior as hypocrisy, they will re-examine their emotional expression according to the information conveyed by the leader’s hypocrisy and then reduce the emotional resources invested in the relationship based on reciprocity. Hence, our research hypothesis:

Hypothesis 1b: Leader hypocrisy negatively influences employees’ affect-based trust.

Trust and Voice Behavior

As the fundamental component of organizations, employees are closely connected with organization improvements in multiple essential aspects, such as researching information, identifying problems, and exploring opportunities.^57,58 Their insights into daily work are significant in improving organizational processes, operational efficiency, and avoiding potential risks.⁵⁶ In order to improve organizational performance, employees’ upward voice behavior (ie, employees actively offer suggestions or opinions to improve the organization’s current condition) has become a bridge linking employees’ opinions to the organization.¹ To be more specific, with the help of the leader’s organizational influence, employees could express their constructive ideas or thoughts upward through voice behavior and finally realize their desire to improve the organization by communicating with the leader. From another perspective, employee voices help leaders further optimize work processes and promote overall organizational growth. However, although speaking out could benefit both employees and the organization, employees engaging in voice behavior must take certain risks, as such challenging behavior reflects changes in the status quo.²³ For example, an employee’s outspokenness may be rejected or treated negatively by the leader, leading to conflict between the two.⁵⁸ In order to avoid potential risks, employees tend to evaluate the risks before making a voice behavior, including assessing the current situation to confirm whether the suggestion could pose threats to them and whether the voice could change the organization.⁵⁹ Thus, the expected reaction of the leader has a significant impact on employees’ voice behavior.

According to social cognitive theory, psychological factors (eg, individual expectations, beliefs, intentions, goals) determine individuals’ behavior.²⁵ Therefore, our study infers that the reduction of trust in leaders will be further reflected in employees’ voice behavior. From one perspective, a lower cognition-based trust between the follower and the leader reduces individuals’ assessment of the safety and effectiveness of voice behavior even if followers realize issues at the organization. In the meantime, low cognition-based trust reflects that employees have doubts about their leaders’ judgment and decision-making ability. Employees do not believe the leader’s judgment of their voice and are worried that the leader may even see the voice as provoking behavior that challenges their authority. Therefore, employees’ voice behavior will be significantly reduced under the low cognition-based trust as the consideration of avoiding risks. From another perspective, the reduction of affect-based trust also reduces the willingness of employees to speak up. The reduction of emotional dependence of employees lends to skeptical attitudes toward the leader. Employees with low levels of affect-based trust are often unwilling to invest additional personal resources or make extra efforts to solve problems for their leaders. Moreover, to avoid damaging relationships with the leader, employees tend to reduce risky behaviors in their daily work and vent their dissatisfaction through “inaction”, resulting in a decline in voice behavior.

Meanwhile, the accumulated feeling of untrust from cognition and affect could break employees’ positive evaluations of their jobs and organization, leading them to express dissatisfaction by reducing proactive behavior and increasing unethical behavior.⁶⁰ Voice behavior belongs to a pro-organizational citizenship behavior with a characteristic of self-determining, which reflects that the individual’s evaluation of the organization essentially determines their voice behavior.²³ Hence, the constantly increasing negative thoughts toward the work, leader, and organization seriously damage the voice intention and then reduce the voice behavior.

Further, based on the principle of “environmental stimulus - mental reaction - behavior”, our research hypothesizes that leader hypocrisy will decrease employees’ voice behavior by reducing cognition-based and affect-based trust. To be more specific, employees who perceive the hypocrisy of a leader may doubt his or her reliability and trustworthiness and reduce their trust in their leaders. Such low trust negatively affects employees’ willingness to engage in pro-organizational behavior, such as active voice behavior. Hence, we hypothesize:

Hypothesis 2a: Cognition-based trust plays a mediate effect in leader hypocrisy and voice behavior.

The Moderate Effect of Moral Identity

The moral identity reflects the different personal degrees of importance toward morality, which is a stable self-recognition based on specific moral characteristics.⁶¹ Past research suggests that moral identity is a personality trait regarding moral orientation closely related to an individual’s cognition of the external environment.⁶¹ People with high moral identity tend to internalize altruistic traits such as honesty, helpfulness, and friendliness into their self-concept.² Those people not only self-behave based on those moral characteristics but also see those moral characteristics as judgment criteria for others’ behavior. For instance, Taylor et al⁶² revealed that people with high moral identity are more likely to negatively evaluate a leader’s abusive management, which causes more negative emotions. On the contrary, people with low moral identities are less sensitive to immoral behavior and less sensitive to whether others behave morally or not.

Based on social cognition theory, people with different characteristics react differently when facing the same event.²⁵ Following the theory, our study believes moral identity could moderate the relationship between leader hypocrisy and cognition-based and affect-based trust. First, people with different levels of moral identity have different levels of sensitivity to morality.⁶¹ Individuals make a moral judgment according to their degree of moral identification. Those with high moral identity regard honesty, trustworthiness, and other moral characteristics as critical personal qualities and tend to judge behaviors in moral “gray areas” as immoral. In contrast, individuals with low moral identity do not emphasize the importance of moral quality and are less sensitive to the leader’s hypocritical behavior. Hence, employees with high moral identity are more likely to identify underlying dishonesty in the leader’s hypocrisy and realize the moral issues from inconsistency in words and deeds, leading to lower cognition-based and affect-based trust toward their leader. On the contrary, employees with low moral identity are not sensitive to the leader’s hypocrisy since they do not care whether the leader’s behavior conforms to the moral standard, resulting in more minor changes in cognition-based and affect-based trust. Second, when there is a value inconsistency between two parties (eg, moral identity), there will be less recognition.⁶³ Leader hypocrisy runs counter to the value proposition of moral identity and is likely to be seen as a violation of the code of ethics. Therefore, when faced with leader hypocrisy, employees who attach great importance to moral concepts are likelier to perceive this behavior as a serious breach of ethical standards and to feel more vigorous disapproval of their hypocritical behavior. Conversely, employees with low levels of moral identity have lower compliance with moral behavior than those with a high level of moral identity. In facing moral detachment behaviors such as deviation, evasion, and uncivilization, employees do not produce detachment but may further relax self-moral constraints. Hence, employees with low moral identity have higher acceptance of leader hypocrisy than those with high moral identity. Hence, we hypothesize:

Hypothesis 3a: Moral identity positively moderates the relationship between leader hypocrisy and cognition-based trust.

Method

Sample and Data Collection

We conducted a three-wave survey in a large Chinese corporation to test the hypothesized model. We first get in touch with the HR manager of this company. After discussing with him the study details and the future steps of the survey, we obtained the company staff’s list (which includes the name, work number, and department). Among these employees, we randomly selected 620 of them to participate in the subsequent questionnaire survey. Meanwhile, we provided each employee with a specific number to match the questionnaires they provided in each survey wave. Before conducting our survey, we provided all the participants with a fifteen-minute meeting to introduce our research’s confidentiality, authenticity, and academic purpose. Notably, as we need to gather sensitive information from employees, we committed to strictly protecting their answers from being seen by their leaders or colleagues by instantly sealing the completed questionnaires in Ziploc bags. Meanwhile, we also made several adjustments to our research process, for example, requiring employees in the same department to complete the questionnaire in batches. After the above steps, 562 employees were willing to participate in this survey.

The questionnaire survey was conducted from October 2022 to December 2022. We kept a three-week time lag between each wave of data collection to avoid common methodbias issues. In the first wave, employees were required to rate their moral identity and their current leaders’ hypocrisy. Meanwhile, employees were required to provide their basic information (gender, age, educational background, tenure, and the years they worked with the current leader). We distributed questionnaires to 562 employees, and 413 of them returned their questionnaires (the response rate is 73.5%). Three weeks later, employees completed measures of cognition-based and affect-based trust. In this wave, we received 386 finished questionnaires (the response rate is 93.4%). In the third wave of data collection, participants were required to finish the measure of voice behavior, and 377 of them provided their answers (the response rate is 97.7%). After removing invalid questionnaires containing missing answers, we finally collected 360 valid questionnaires, a 64.1% return rate of 562 questionnaires initially distributed.

Our samples contain 49.2% men and 50.8% women. There were 12.2% between 18–25 years old, 27.2% between 26–35 years old, 36–45 years old accounted for 39.2%, 19.2% were between 46–55 years old, and 2.2% were older than 56 years old. Regarding educational background, our samples included 3.3% in junior high school and below, 15.0% in high school and technical secondary school, 30.8% in junior college, 41.4% in undergraduate, and 9.4% in master’s degree and above. In terms of tenure, there were 13.6% worked in the company for less than one year, 19.8% worked for 1–2 years, 26.9% worked for 2–3 years, 26.1% worked for 3–4 years, 11.7% worked for 5–10 years, 1.9% worked for more than ten years. From the perspective of years working with the current leader, our sample contained 18.6% for less than one year, 36.9% for 1–2 years, 30.6% for 2–3 years, 9.2% for 3–4 years, 4.2% for 5 −10 years, and 0.6% for more than ten years.

Measurement

All measurements used in our study are mature scales developed by scholars. Our study followed the “reverse translation” procedure to translate the English scales into Chinese scales that are suitable for the context of Chinese enterprises in order to ensure semantic consistency between Chinese and English. We accessed all questionnaire items using a five-point Likert scale from 1, “very inconsistent”, to 5, “very consistent”.

Leader hypocrisy: Consistent with Greenbaum, Mawritz, and Piccolo,²⁴ we tested leader hypocrisy by adapting the Behavioral Consistency Scale developed by Dineen, Lewicki, and Tomlinson.⁶⁴ The scale includes four items. An example item is “My supervisor tells us to follow the rules but doesn’t follow them himself or herself”.

Cognition-based trust: We measured cognition-based trust by applying the scale developed by Ng and Chua.⁶⁵ The scale includes four items. An example item is “You can rely on them to do a major portion of the group work”.

Affect-based trust: We applied the scale developed by Ng and Chua⁶⁵ to measure affect-based trust. The scale includes four items, and the example item is “You can talk freely to them about your difficulties and know that they will want to listen”.

Voice behavior: We measured employees’ voice behavior by applying the scale developed by Van Dyne and LePine.⁶⁶ The scale contains six items, and the example item is “I speak up and encourage others in this group to get involved in issues that affect the group”.

Moral identity: We measured moral identity by applying the scale developed by Reiss and Mitra.⁶⁷ The scale includes five items such as “It would make me feel good to be a person who is compassionate, fair, friendly, generous, hardworking, helpful, honest, and kind”.

Control variables: Taking reference from current studies, our research controlled employees’ gender, age, education degree, and tenure. In addition, we controlled working years with the leader due to its influence on employees’ attitudes and behavior.

Result

Validity & Reliability

We test the reliability and validity of the scales used in the study. As shown in Table 1, Cronbach’s α coefficients of the five core variables are between 0.880–0.956, all greater than 0.8, and the combined reliability (CR) value is also greater than 0.9 (see detail in Table 1). The result indicates that the questionnaire has good internal consistency. In terms of validity, the KMO value of each variable is greater than 0.8, and the AVE value is above 0.6, indicating that the questionnaire has high convergent validity.

Table 1 Testing of Reliability and Validity

Confirmatory Factor Analysis

In order to further verify the discriminant validity of the variables, we conducted a confirmatory factor analysis on leadership hypocrisy, cognition-based trust, affect-based trust, voice behavior, and moral identity. As shown in Table 2, compared with other models, the five-factor model had the best fitting effect (χ2 = 401.95, χ2/df = 1.83, RMSEA = 0.05, CFI = 0.97, TLI = 0.97, SRMR = 0.04). The results indicated that the variables in this study had good discriminant validity.

Table 2 Confirmatory Factor Analysis

Common Method Bias Test

We applied the Harman single-factor test method to test whether there is a common method bias, given that the data used in our study were all self-reported. The results showed that five factors with eigenvalues greater than one were obtained without rotation, and the first factor explained 36.17% of the variance, which was less than 40%.⁶⁸ The result showed no serious method bias problem in the study.

Descriptive Analysis

Table 3 includes each variable’s mean, standard deviation, and correlation coefficient (See detail in Table 3). The results showed that leadership hypocrisy was significantly negatively correlated with cognition-based trust (r = −0.24, p < 0.01), as well as affect-based trust (r = −0.25, p < 0.01). Cognition-based trust was significantly positively correlated with voice behavior (r = 0.31, p < 0.01), and affect-based trust was significantly positively correlated with voice behavior (r = 0.40, p < 0.01).

Table 3 Descriptive Statistic

Hypothesis Testing

Direct effect and the mediating effect. Hypothesis 1 indicates that leadership hypocrisy reduces employees’ cognition-based trust (H1a) and affect-based trust (H1b). To test Hypothesis 1, we conducted regression analysis on the relationship between leader hypocrisy, cognition-based trust, and affect-based trust by controlling for gender, age, education, years of work, and years of working with the current leader. The result is shown in Table 4 (see details in Table 4). It can be seen from Table 4 that leadership hypocrisy has a significant negative effect on cognition-based trust (b = −0.25, 95% C1 = [−0.37, −0.14], excluding 0) and significantly negatively influences affect-based trust (b = −0.31, 95% C1 = [−0.45, −0.17], excluding 0). Therefore, Hypotheses 1a and 1b were supported.

Table 4 Analysis Result of Direct Effect & Mediating Effect

Hypothesis 2a indicates that cognition-based trust mediates the relationship between leadership hypocrisy and employee voice behavior. Table 4 shows that the mediating effect of cognition-based trust between leadership hypocrisy and voice behavior is b = −0.07, 95% C1 = [−0.11, −0.03], excluding 0. Therefore, Hypothesis 2a is supported.

Hypothesis 2b indicates that affect-based trust mediates the relationship between leadership hypocrisy and employee voice behavior. The results in Table 4 show that the mediating effect coefficient of affect-based trust between leadership hypocrisy and voice behavior is b = −0.09, 95% C1 = [−0.15, −0.05], excluding 0. Therefore, Hypothesis 2b is supported.2.

Moderating effect test. Hypothesis 3a predicts that moral identity positively moderates the relationship between leadership hypocrisy and cognition-based trust. According to Table 5, the interaction term of leadership hypocrisy and moral identity significantly negatively affects cognition-based trust (b = −0.13, 95% C1 = [−0.24, −0.01], excluding 0 (see details in Table 5). Considering that leadership hypocrisy negatively affects cognition-based trust, moral identity positively moderates the impact of leadership hypocrisy on cognition-based trust. In order to demonstrate the moderating effect of moral identity more clearly, our study utilized the suggestions of Aiken et al⁶⁹ to draw the relationship between hypocrisy and cognition-based trust when moral identity is one standard deviation above and below the mean. As shown in Figure 2, when moral identity is high, leadership hypocrisy has a substantial impact on cognition-based trust (b = −0.28, 95% C1 = [−0.42, −0.13], excluding 0); when moral identity is low, the effect of leadership hypocrisy on cognition-based trust is weak and insignificant (b = −0.08, 95% C1 = [−0.20, 0.12], 0 inclusive). Therefore, Hypothesis 3a is supported.

Table 5 Summary of Moderating Effect of Moral Identity on the Relationship Between Leader Hypocrisy and Cognition-Based Trust & Affect-Based Trust

Figure 2 The moderating effect of moral identity on the relationship between leader hypocrisy and cognition-based trust.

Hypothesis 3b predicts that moral identification positively moderates the relationship between leadership hypocrisy and affect-based trust. Table 5 shows leadership hypocrisy and moral identity interaction significantly negatively affect affect-based trust (b = −0.17, 95% C1 = [−0.31, −0.02], excluding 0). Similarly, considering that leadership hypocrisy significantly negatively affects affect-based trust, moral identity here positively moderates the impact of leadership hypocrisy on affect-based trust. In order to show the moderating effect of moral identity more clearly, our study draws the relationship between hypocrisy and affect-based trust when moral identity is one standard deviation above and below the mean. As shown in Figure 3, when moral identity is high, leadership hypocrisy strongly and significantly impacts affect-based trust (b = −0.35, 95% C1 = [−0.53, −0.16], excluding 0). When moral identity is low, the effect of leadership hypocrisy on affect-based trust is weak and insignificant (b = −0.04, 95% C1 = [−0.24, 0.16], 0 inclusive). Therefore, Hypothesis 3b is supported.

Figure 3 The moderating effect of moral identity on the relationship between leader hypocrisy and affect-based trust.

Discussion

This study substantially advances our understanding of voice behavior within organizations, particularly focusing on the negative consequences of leader hypocrisy. Previous research primarily examined how voice behavior influences certain leadership styles,^33,70,71 limited research in deeply exploring the impact of specific leader traits,^72,73 especially the negative traits. Considering that “Nobody’s perfect”, the discrepancies between leaders’ words and behavior which may be attributed to leaders’ limited energy can also generate adverse reactions among organizational members. This study aims to fill the critical gap in understanding how employees react to such inconsistencies in leadership behavior and the subsequent effects on their voice behavior.

Utilizing social cognitive theory, our research confirms and expands on existing knowledge about the negative effects of leader hypocrisy on employee voice behavior. We explore the complex mechanisms and situational factor that govern this relationship. We identify the dual mediation of cognition-based and affect-based trust as crucial. This dual-pathway mediation provides a detailed perspective on how employees’ perceptions of leader hypocrisy lead to reduced voice behavior. The decline in cognition-based trust indicates doubts about the leader’s competence and trustworthiness, while a reduction in affect-based trust points to weakened emotional connections and respect for the leader. Our study also sheds light on the significant moderating role of moral identity in this dynamic. We found employees with a strong moral identity are more negatively affected by leader hypocrisy, likely due to their higher commitment to ethical consistency and integrity. Such individuals might perceive hypocritical behavior as particularly jarring, resulting in a more substantial erosion of trust and increased reluctance to share their views and suggestions. By doing so, our exploration into the effects of leader hypocrisy on employees’ willingness to speak up and their cognitive processes offers a more comprehensive and profound insight into how negative leadership characteristics affect the workplace environment.⁷⁴

Theoretical Implications

First, our study extends the voice behavior literature by examining the effect of “undoing” leader behavior on voice. Existing research primarily focuses on the influence of leaders’ macro styles and traits on employee voice behavior, such as transformational leadership,⁷⁵ authentic leadership,⁷⁶ inclusive leadership,⁷⁷ personality traits,⁷⁸ and leader humility.⁷⁹ However, as organizational practice shows, even leaders with the same style may exhibit different behaviors.⁸⁰ This implies that current research may struggle to accurately explain the impact of leaders on employee voice behavior. Therefore, it is essential to conduct a more detailed and in-depth exploration of the antecedents of voice behavior from a micro perspective, an aspect overlooked by existing research. In light of this, we examined the impact of leader hypocrisy (considered a widespread phenomenon in organizational practice and having significant effects on followers’ evaluations and behaviors²⁸) on employee voice behavior. Our study indicated that leader hypocrisy (word–deed misalignment) adversely affected employees’ cognition-based and affect-based trust in leaders, thereby inhibiting their voice behavior. Consequently, we enriched the literature on voice behavior. At the same time, we have also further revealed the dangers of leaders’ “fake actions”, responding to the call by Wang et al⁸¹ for more in-depth exploration into the consequences of poor leadership behavior.

Secondly, although leader hypocrisy is a phenomenon in the workplace with negative consequences, there are limited studies exploring the consequences caused by it. Most research investigates the effect of leader hypocrisy on employees’ behavior by examining the mediating role of employees’ psychological state,^24,81 the current impact pathways are relatively homogenous and limited. Our study expands the scarce literature on leader hypocrisy by examining its effect on employees’ voice behavior through a dual pathway – cognition-based and affect-based trust. By doing so, we reveal the effects of leader hypocrisy on followers’ evaluation of the leader and its influence on voice behavior, thus providing new research perspectives for future studies. Meanwhile, our research response to Ilsev and Aydin’s⁷⁴ calls to conduct more comprehensive research on the emotional, attitudinal, and behavioral consequences of a leader’s word-deed misalignment.

Finally, from the perspective of employees’ traits (morality), we examined the moderating role of moral identity in the impact of leader hypocrisy on employee voice behavior, thus supplementing the boundary conditions of the antecedents (leader hypocrisy) of voice behavior from a novel perspective. Leader hypocrisy typically occurs during the interaction between leaders and employees and contradicts the moral notion of consistency in words and behavior. As emphasized in social cognition theory, an individual’s beliefs and traits influence their attitudes and behaviors towards the external environment. Therefore, we empirically tested the moderating effect of moral identity in the relationship between leader hypocrisy and employee voice behavior. That is, for employees with a high level of moral identity, the negative impact of leader hypocrisy on their cognition and affect-based trust was strengthened. In doing so, we enriched the literature on leader hypocrisy from a moral perspective.

Practical Implications

This study offers valuable practical implications for leaders and organizations. First, our findings underscore the necessity for leaders to align their behaviors with words, which is crucial in maintaining employee trust. Therefore, we recommend that organizations and managers proactively, promptly, and transparently address any discrepancies between their words and actions to mitigate the negative impacts of leader hypocrisy. Specifically, it is paramount for organizations to develop clear behavioral guidelines for managers, emphasizing honesty, transparency, and consistency. Then, leaders should be selected and developed based on these guidelines. For instance, leaders should also be encouraged to regularly engage in self-reflection through personal journals, reflective meetings, or professional coaching to help leaders identify and address their shortcomings consistently. Organizations should regularly review leaders’ behaviors to ensure their actions align with organizational guidelines through acquiring feedback from subordinates, peer reviews, superior evaluations, or performance assessments, and reward or penalize accordingly. These systematic management measures help to reduce leader hypocrisy.

Second, as our research results indicate, leader hypocrisy severely impacts employee voice behavior. Therefore, we propose a key strategy: developing diverse channels for employee voice to counteract the negative effects of perceived leader hypocrisy. This can be achieved by establishing a feedback platform and creating an open communication environment. Organizations can create an anonymous feedback platform, such as online surveys, internal forums, complaint, and suggestion boxes, empowering employees to voice openly and safely. It is also crucial to encourage them to exercise this power, necessitating the creation of an open communication environment. For instance, organizations should regularly inform employees about whether their voices are endorsed, and recognize and reward those who provide valuable feedback. These measures can foster an open and honest organizational culture. Thus, the combination of institution and culture enhances employees’ trust in the organization and their leaders, reduces the negative impact of leader hypocrisy, and thereby promotes employee voice behavior.

Our third practical implication extends from the first two. We identified the significance of moral identity in moderating the impact of leader hypocrisy, prompting us to contemplate ethical issues in management practices. Hence, we propose that organizations should strengthen the establishment of ethical systems. The foundation of building an ethical system is a detailed code of ethics, which clearly defines the behavioral standards expected by the organization, including consistency, integrity, and fairness. To ensure the implementation of the ethical code, organizations should also set up an Ethics and Morality Committee responsible for employees’ ethical modules. This includes regular ethical performance assessments and behavioral supervision of leaders and employees, as well as providing consultation, guidance, and assistance in ethical matters for employees. Additionally, leaders should also become moral examples. For instance, they should publicly demonstrate their commitment to the organization’s ethical code and embody moral principles in daily decision-making.

Limitations and Future Research

While we believe our research makes some contributions, it is not without limitations that open up future research opportunities. First, the questionnaires in the study are all employee self-reported, which is considered as a single sample source. Although we tested Harman’s single factor, and the result showed that the problem of common method bias was not significant, we encourage future researchers to try to collect data from multiple sources. For example, employee voice behavior can be measured by their leaders rather than employees’ self-reporting. Second, even though we collect questionnaires in three waves, our data remain essentially cross-sectional, which limits our judgment of causality. Future research is encouraged to revalidate our research results with other research methods (eg, experiments and experience sampling methods). Third, our study regards leadership hypocrisy as a general negative variable and thus concludes that after employees discover leadership hypocrisy, both cognition-based trust and emotional trust in leaders will decline. However, if employees find that the leader’s hypocrisy is justified or there are reasons beyond the leader’s control, employees might be more tolerant and understand the leader.⁴² Therefore, follow-up research can further explore the employees’ emotions, attitudes, and subsequent behavioral changes toward the leader based on different reasons for leader’s hypocrisy.