Psychological Reactance to Leader Moral Hypocrisy

Abstract

Drawing on early work on ethical leadership, we argue that when leaders engage in leader moral hypocrisy (i.e., ethical promotion without ethical demonstration), followers can experience psychological reactance—a negative response to a perceived restriction of freedom—which can have negative downstream consequences. In a survey of employee–manager dyads (study 1), we demonstrate that leader moral hypocrisy is positively associated with follower psychological reactance, which increases follower deviance. In two subsequent laboratory experiments, we find similar patterns of results (study 2) and explore potential alternative mechanisms (study 3). We demonstrate in a final experiment with working adults that the relationship between leader moral hypocrisy and psychological reactance is partly explained by increased perceptions of a leader’s use of power (study 4). We discuss the implications of our findings and advocate for further understanding of the risks associated with psychological reactance in response to leaders and other workplace situations.

There is general consensus among scholars that followers’ reactions to ethical leadership are largely positive (Brown & Treviño, 2006; Brown, Treviño, & Harrison, 2005; Den Hartog, 2015; Mayer, Aquino, Greenbaum, & Kuenzi, 2012; Mayer, Kuenzi, Greenbaum, Bardes, & Salvador, 2009). Ethical leadership tends to make employees more prosocial (Ng & Feldman, 2015), less deviant (Schaubroeck et al., 2012), more satisfied (Bedi, Alpaslan, & Green, 2016), and less likely to want to quit (Ng & Feldman, 2015). According to theory on ethical leadership, these outcomes stem from employee reactions to two key sets of leader behaviors. First, ethical leaders engage in ethical demonstration, reflective of a moral person who is seen as an example of integrity and normatively appropriate conduct in the organization. Second, ethical leaders engage in ethical promotion, reflective of a moral manager who encourages normatively appropriate conduct among followers (Brown et al., 2005). Research on ethical leadership, therefore, focuses primarily on followers’ reactions to leaders who engage in both ethical demonstration and ethical promotion (Den Hartog, 2015).

However, much less research has empirically examined followers’ reactions to leader moral hypocrisy—when leaders actively promote ethics within their organizations but fail to demonstrate ethical behavior themselves (for a rare exception, see Bush, Welsh, Baer, & Waldman, 2021). ¹ In their initial formulation of ethical leadership, Treviño, Hartman, and Brown (2000) theorized that followers would disregard the ethical promotion of morally hypocritical leaders. But we argue that reactions to leader moral hypocrisy may go one step further, not only leading followers to disregard or withdraw from their leaders’ ethical promotion but also motivating them to behave in ways that directly oppose such promotion through unethical behaviors that are actively harmful to the organization. To explain such behavior, we propose and examine psychological reactance as an important phenomenon.

Psychological reactance is a strong negative reaction to a perceived restriction of freedom that motivates people to actively resist behaviors that they feel are being forced upon them (S. S. Brehm & Brehm, 1981). To better understand psychological reactance to leader moral hypocrisy in the workplace, we explore both the perceptions followers have of their leaders that provoke psychological reactance and the potential unethical behaviors in which followers engage as a response to such reactance. Specifically, we argue that because morally hypocritical leaders forfeit their moral authority (Isserow & Klein, 2017), their attempts to promote ethics to followers will be perceived as a double standard and received as an undue exercise of power. We further argue, and empirically demonstrate, that followers’ perceptions of their morally hypocritical leaders’ exercise of power can evoke psychological reactance, leading to negative downstream consequences.

With this research, we offer new theoretical and empirical insights into followers’ reactions to leader moral hypocrisy. We demonstrate the importance of understanding the impact that psychological reactance may have in pushing followers from simply ignoring their leaders’ directives to actively working to counter such directives. Leaders simply espousing ethics and telling their employees to be ethical will not suffice; they need to actually be ethical themselves to avoid backlash from followers. In addition to further exploring the consequences of leader moral hypocrisy, we extend management research by offering insight into the role that psychological reactance can play in the workplace.

LEADER MORAL HYPOCRISY

Ethical leadership has been defined as a combination of 1) the “demonstration of normatively appropriate conduct through personal actions and interpersonal relationships” and 2) the “promotion of such conduct to followers through two-way communication, reinforcement, and decision-making” (Brown et al., 2005: 120, italics added). Treviño et al. (2000) acknowledged that ethical demonstration and ethical promotion were independent of one another, presenting a 2 × 2 model of four quadrants of leadership related to ethics: 1) an ethical leader who both promotes and demonstrates ethical behaviors, 2) an ethically mute leader who demonstrates but does not promote ethical behaviors 3), an unethical leader who does not demonstrate or promote ethical behaviors, and 4) a hypocritical leader who promotes but does not demonstrate ethical behaviors.

Although much research has examined the ethical leader quadrant (Den Hartog, 2015), much less work has focused on the hypocritical leader quadrant, what we term leader moral hypocrisy to emphasize that the hypocrisy is specifically related to moral or ethical issues. How do employees react, for example, to leaders who promote the importance of respect and inclusion at work but have been known to engage in discriminatory practices themselves, or to leaders who mandate that bribes must not be paid to win contracts from foreign governments but who are notorious for resorting to bribes when the “price is right”? From work on hypocrisy in general, it is unsurprising that followers often react negatively to leaders who are hypocritical. Much of this research has demonstrated that followers tend to withdraw from their hypocritical leaders (Bush et al., 2021; Scheidler, Edinger-Schons, Spanjol, & Wieseke, 2019; Simons, Leroy, Collewaert, & Masschelein, 2015; Xu, Loi, & Lam, 2015). For example, Greenbaum, Mawritz, and Piccolo (2015) found that leader hypocrisy was associated with follower turnover intentions. However, from the extant literature, it is unclear why leader hypocrisy—and leader moral hypocrisy, specifically—might lead employees not only to disregard their leaders’ directives but to purposefully oppose them by engaging in deviant behavior. We argue that when people feel that ethics and morality are being “pushed” on them—especially by a leader who lacks moral integrity themselves—they can be motivated to actively work against their leaders’ ethical promotion owing to the experience of psychological reactance.

PSYCHOLOGICAL REACTANCE TO LEADER MORAL HYPOCRISY

According to reactance theory, psychological reactance involves a motivation directed at reclaiming or maintaining a sense of freedom over a given choice (J. W. Brehm, 1966). For example, Pavey, Churchill, and Sparks (2022: 2) explain that psychological reactance “subsequently results in a person responding to a persuasive attempt in the opposite way to that intended, to reinstate the threatened freedom.” Psychological reactance thus entails both a sense of restricted freedom in a particular domain and a desire to reassert one’s autonomy in that domain. One of the most common ways psychological reactance is provoked is through external pressure from others (Worchel & Brehm, 1971). For example, when individuals receive unsolicited advice, they often experience reactance to that advice, ultimately selecting an option that reflects the opposite of the advice they received (Fitzsimons & Lehmann, 2004). Pressure from authority figures seems especially likely to elicit psychological reactance and its behavioral manifestations. For example, Pennebaker and Sanders (1976) placed signs from university security officials prohibiting graffiti on the walls of toilet stalls in a library, resulting in increased instances of graffiti. Public health initiatives designed to limit smoking options (e.g., government controls to make smoking less accessible through increased taxes) ended up being positively related to rates of smoking (Wiium, Aarø, & Hetland, 2009). It has even been shown that when people feel that someone is pushing them to agree with a certain opinion on an issue, they are more likely to disagree with that opinion (Carver, 1977; Silvia, 2006).

With respect to morality and ethics in the workplace, it makes intuitive sense that some followers might experience psychological reactance to ethical leadership if their ethical standards differ from those they see as being promoted and enforced by their leader. Perhaps less intuitive, however, is that psychological reactance may occur even when followers’ ethical standards are consistent with those their hypocritical leader is promoting. In such cases, we argue that leader moral hypocrisy will trigger psychological reactance among followers because it will be perceived largely as a coercive exercise of power, which can be experienced as a restriction of freedom (Bierstedt, 1950; Pavey et al., 2022). At the heart of our argument that leader moral hypocrisy can provoke followers’ psychological reactance, therefore, lies the important role of followers’ perceptions of their leaders’ coercive use of power.

First, we contend that ethics-related influence attempts made by morally hypocritical leaders will be perceived as a coercive exercise of formal power because such leaders forfeit their moral authority. Whereas a leader’s formal authority is based largely on their position of power within an organization, their moral authority refers to their potential to influence the morality of others based on perceptions that they, the leader, are moral themselves (see Hoppner & Vadakkepatt, 2019; White, 1996). However, as Isserow and Klein (2017: 201) explain, hypocrisy undermines a person’s moral authority:

It is not merely that we doubt the hypocrite’s integrity, or question their moral compass. We also tend to think that they are no longer warranted … in condemning certain kinds of behaviour.… Their actions are taken to suggest that the esteem and deference extended to them was not deserved.

Thus, whereas ethical leaders can rely on their moral authority to appeal to followers’ values on a more volitional level, morally hypocritical leaders must rely more on their formal power to coerce followers on moral matters. Failing to practice what they preach, the attempts of morally hypocritical leaders to promote ethics in the workplace will be perceived as a double standard and feel more like a “power play.” Devoid of moral authority, the ethical promotion of morally hypocritical leaders will be more likely perceived by followers as a coercive exercise of formal power based on positional authority.

Second, once followers perceive the ethical promotion of their morally hypocritical leaders as an exercise of power, we posit that they will be more likely to experience psychological reactance. Especially given the hierarchies inherent in most organizations, any coercive tactics followers perceive from their leader will feel like a threat to their autonomy and freedom. Such feelings may increase the salience of a follower’s low-power position, which is psychologically associated with less control and a constraint on one’s ability to act (Lammers, Stoker, Rink, & Galinsky, 2016; Magee & Galinsky, 2008). Indeed, research has demonstrated that low power “conflicts with the basic human need for control and autonomy” (Guinote, 2017: 357)—which, again, can exacerbate psychological reactance. Followers may even resent feeling forced to act in ways that they would typically have wanted to act of their own volition. As several decades of research on self-determination theory have shown, people are less motivated and less likely to engage in behaviors when they feel forced, even when those behaviors had previously been valued and even enjoyed (Deci & Ryan, 1985; Moller, Ryan, & Deci, 2006).

In sum, drawing from reactance theory and research on ethical leadership, we argue that leader moral hypocrisy (i.e., when leaders promote but do not demonstrate ethics) can evoke psychological reactance among followers. We further expect that this relationship is explained, at least partly, by increased perceptions of a leader’s use of power. Thus we formally predict the following:

Hypothesis 1a: Leader moral hypocrisy is positively associated with psychological reactance.

Hypothesis 1b: The positive relationship between leader moral hypocrisy and psychological reactance is mediated by followers’ perceptions of a leader’s use of power.

DOWNSTREAM CONSEQUENCES OF PSYCHOLOGICAL REACTANCE

Although freedoms that are seen as restricted can be reestablished in many ways, the most direct way a person can remedy the psychological reactance they feel is through “engagement in the threatened behavior” (S. S. Brehm, 1981: 939). In response to experiencing psychological reactance, people can actively resist restrictions they feel are being imposed upon them by behaving in ways that directly contrast with the behaviors that a persuader is trying to elicit—what S. S. Brehm and Brehm (1981) describe as a boomerang effect.

With respect to leader moral hypocrisy, we assert that when followers experience psychological reactance, rather than simply disregarding their leader’s ethical directives, they might engage in even more unethical and deviant workplace behavior—as they seek to reassert their freedom and autonomy. As reactance theory (J. W. Brehm, 1966; S. S. Brehm & Brehm, 1981) would predict, employees who feel that certain behaviors and restrictions are being imposed upon them will experience a desire to resist such impositions. Therefore, rather than simply withdrawing from or disregarding the ethical promotion of a hypocritical leader—as Treviño et al. (2000) propose—followers experiencing psychological reactance may be motivated to engage in morally questionable behavior more directly in an attempt to reassert their freedom. We therefore predict that increased workplace deviance can be a downstream consequence of a follower’s psychological reactance in response to leader moral hypocrisy (see Figure 1 for our full theoretical model):

Hypothesis 2: Leader moral hypocrisy has a positive indirect effect on the workplace deviance of followers through psychological reactance as a mediator.

Figure 1: Theoretical Model

OVERVIEW OF STUDIES

Across four studies, we investigated the relationship between leader moral hypocrisy and follower psychological reactance. In studies 1 and 2, we examined leader moral hypocrisy by examining the interaction effect of ethical promotion by ethical demonstration, which we measured (in a field survey of paired employee–manager dyads; study 1) and manipulated (in a scenario-based experiment; study 2) separately. Doing so allowed us to compare the effect of all four of Treviño et al.’s (2000) ethical leadership quadrants on psychological reactance, as well as on employee deviance as a downstream consequence. In studies 3 and 4, we narrowed our focus and conducted scenario-based experiments that manipulated leader moral hypocrisy (as a single construct) to rule out alternative mediators explaining its effect on deviant behavior (study 3) and to test whether followers’ perceptions of their leader’s use of power serves as a mechanism explaining the effect of leader moral hypocrisy on follower psychological reactance (study 4). Taken together, our four studies provide initial evidence that psychological reactance may be a consequential response to leader moral hypocrisy.

STUDY 1

In study 1, we examined the effect of leader moral hypocrisy on followers’ psychological reactance at work and the downstream consequences on employee deviance. To do so, we collected data from both employees and their managers across a wide variety of organizations and industries.

Sample

We requested survey responses from employee–manager dyads through an opt-in, online panel of individuals willing to participate in academic surveys for payment—a service that other management researchers have utilized (e.g., Ng & Feldman, 2013; Piccolo & Colquitt, 2006). Out of 440 requests, 406 individuals (210 employees and 196 managers) submitted a survey online. After matching employees and managers, eighteen employees had responded to the survey whose managers had not, and four managers responded to the survey whose employees had not. We removed these individuals’ responses from our data set, resulting in a sample of 192 dyads for which we had responses from both employee and manager.

Seventy-three percent of managers were male, and their average age was 38.37 years (SD = 7.51). Of these managers, 83.76 percent identified as white/Caucasian, 4.71 percent as Black/African American/African, 3.14 percent as Asian, 4.19 percent as Hispanic, and 4.20 percent as another race. Seventy-six percent of employees were male, and their average age was 35.39 years (SD = 6.58). Of these employees, 76.43 percent identified as white/Caucasian, 6.80 percent as Black/African American/African, 6.28 percent as Asian, 6.28 percent as Hispanic, and 4.19 percent as another race. Employees reported working for their current employer (i.e., organizational tenure) for an average of 8.50 years (SD = 4.76) and being in their current position for an average of 6.55 years (SD = 4.19). Occupations reported by employees varied widely, including customer service representative, graphic designer, computer programmer, laboratory manager, financial specialist, line cook, construction inspector, laborer, office manager, IT manager, consultant, project manager, engineer, receptionist, and so on.

Procedures

All participants were instructed to complete a survey related to their current employment. To investigate the effects of leader moral hypocrisy, we collected measures of both managers’ ethical promotion and their ethical demonstration, with the intent of examining the interaction effect between the two, focusing particularly on the simple slopes associated with high ethical promotion and low ethical demonstration (i.e., leader moral hypocrisy). Moreover, to avoid common method bias and reduce social desirability bias, we asked managers to rate their own ethical promotion, but we asked employees to rate their managers’ ethical demonstration. We also asked managers to rate their employees with respect to workplace deviance, and we asked employees to rate their own level of psychological reactance at work. We asked all participants to provide basic personal and work-related demographic information. Each manager was explicitly informed that their survey responses would never be seen by their employee, and vice versa. After the collection period was complete, we matched data from each manager with their respective employee.

Measures

Leader Moral Hypocrisy

To examine leader moral hypocrisy, we collected separate measures of managers’ ethical promotion and ethical demonstration. Given the challenges associated with difference scores (Edwards, 1994), we did not combine the measures to form a composite score for leader moral hypocrisy; rather, by including the two measures separately in our analysis, we were able to probe the simple slopes of their interaction effect, which allowed us to examine all four quadrants of leadership related to ethics proposed by Treviño et al. (2000)—with high promotion and low demonstration reflecting the leader moral hypocrisy quadrant.

We asked managers to rate the extent to which they engaged in ethical promotion by having them complete the 7-item Ethical Guidance Scale from Kalshoven, Den Hartog, and De Hoogh’s (2011) Ethical Leadership at Work questionnaire. We used this measure because it was written with the intent of assessing the promotion component of ethical leadership specifically, independent of the demonstration component. We asked managers to rate each item with respect to themselves on a scale ranging from 1 (strongly disagree) to 7 (strongly agree). An example item is “Explain what is expected from employees in terms of behaving with integrity.” The items were averaged to form an overall ethical promotion score for each manager (α = .88). Having managers provide self-assessments of ethical promotion allowed us to examine the relationship between ethical promotion and ethical demonstration in a way that was not completely rooted in employees’ perceptions but based on managers’ reports of their own efforts to promote ethics.

We asked employees to rate the ethical demonstration of their managers using Kalshoven et al.’s (2011) 4-item measure of integrity, another dimension of their Ethical Leadership at Work questionnaire. An example item is “keeps his/her promises.” The items were averaged to form an ethical demonstration score for each manager (α = .78).

Psychological Reactance

We used the 11-item measure from Hong and Faedda (1996) to assess the psychological reactance employees experienced in the workplace in general. We adapted each question so that it related to work incidents specifically. Example items include the following: “I resist the attempts of others at work to influence me” and “I become frustrated when I am unable to make free and independent decisions at work.” Participants rated each item on a scale ranging from 1 (completely disagree) to 5 (completely agree). An average of the eleven items was taken to form a composite psychological reactance score for each employee (α = .91).

Deviance at Work

Managers rated their employees’ deviance using Bennett and Robinson’s (2000) Workplace Deviance Scale. Managers indicated the frequency with which their employees engaged in nineteen deviant behaviors, such as “came in late to work without permission”; “made an ethnic, religious, or racial remark at work”; and “taken property from work without permission,” on a 7-point scale ranging from 1 (never) to 7 (daily) (α = .99). ²

Results and Discussion

Means, standard deviations, and correlations among all study variables are displayed in Table 1. As might be expected from previous work, ethical promotion and ethical demonstration were both negatively correlated with employee deviance. Furthermore, psychological reactance was positively correlated with employee deviance.

Table 1: Means, Standard Deviations, and Correlations (Study 1)

Note. EMP = measure reported by the employee. MGR = measure reported by the manager. Scale reliabilities are given in parentheses.

* p < .001.

We tested our hypotheses using moderated-mediation model 7 of Hayes’s (2017) PROCESS macro (version 4) for SPSS version 28. The first path in this model allowed us to explore how the interaction between ethical promotion and ethical demonstration affected psychological reactance (see Table 2). The interaction between ethical promotion and ethical demonstration resulted in a statistically significant effect, b = −0.24, SE = 0.09, p < .01, 95 percent confidence interval (95% CI) [−0.41, −0.07]. An analysis of the simple slopes (Aiken & West, 1991), as displayed graphically in Figure 2, revealed that ethical promotion increases psychological reactance when leader ethical demonstration is lower, b = 0.17, SE = 0.07, p = .02, 95% CI [0.03, 0.31], but not when leader ethical demonstration is higher, b = −0.15, SE = 0.13, p = .22, 95% CI [−0.40, 0.09]. We therefore find support for hypothesis 1a.

Table 2: Interactive Effects of Leader Ethical Promotion and Demonstration on Follower Psychological Reactance and Deviance (Study 1)

* p < .05.

** p < .001.

Figure 2: Interactive Effect of Leader Ethical Promotion and Demonstration on Follower Psychological Reactance (Study 1)

Note. Higher and lower values represent 1 standard deviation above and below the mean of each variable, respectively. Error bars represent standard errors around the mean.

We next examined the interaction effect of ethical promotion and ethical demonstration on deviance through psychological reactance as a mediator. First, we found a significant interaction between ethical promotion and ethical demonstration predicting deviant behavior, b = 0.49, SE = 0.15, p < .01, 95% CI [0.19, 0.79] (see Table 2). Based on the 95 percent confidence intervals, the conditional indirect effects of ethical promotion on deviance through psychological reactance are statistically significant at lower levels of leader ethical demonstration, b = .12, SE = 0.08, 95% CI [0.02, 0.32], but not at higher levels, b = −0.06, SE = 0.08, 95% CI [−0.25, 0.01] (i.e., 1 standard deviation below and above the mean, respectively). We therefore find support for hypothesis 2.

The results of study 1 offer support for our prediction that leader moral hypocrisy (i.e., promoting ethics without visibly demonstrating ethics) comes with a significant risk. Indeed, this style of leadership increased followers’ psychological reactance, which had downstream consequences on employee deviance. These results suggest that leaders would do well to visibly align their behavior with any ethical values they choose to promote among their employees. However, because study 1 relied on correlational data, we were unable to draw strong causal inferences. To address this important limitation, and to further understand the role of psychological reactance in people’s responses to leader moral hypocrisy, we conducted a laboratory experiment.

STUDY 2

Study 2 was designed to allow us to draw stronger causal inferences about the relationship between leader moral hypocrisy and follower psychological reactance. To that end, we employed an experimental paradigm using a vignette in which we directly manipulated a leader’s ethical promotion and demonstration. Doing so allowed us to again examine Treviño et al.’s (2000) four quadrants of ethical leadership: ethical leaders, morally hypocritical leaders (those who engage in leader moral hypocrisy), unethical leaders, and ethically neutral leaders.

Participants

Two hundred ten undergraduate students from a midwestern US university participated in this study for partial course credit. We chose to use a university population for two reasons. First, every person in the sample was earning a business degree, allowing us to ensure that participants understood the scenario we outlined in the study. Second, we were able to expand the generalizability of our findings to show that those with a wide range of work experience might respond similarly to the various types of ethical leadership that we explore. Five participants were excluded from the sample for not passing an attention check, leaving a final sample of 205 (98 women). Participants had an average age of 19.8 years (SD = 1.93), and 48 percent reported being employed. In our sample, 69.11 percent of participants identified as white/Caucasian, 1.90 percent as Black/African American, 12.75 percent as Asian, 11.27 percent as Hispanic, and 4.90 percent as a race or ethnicity other than those listed.

Study Design and Procedure

Upon arriving at a lab, participants were randomly assigned to one of four conditions, in a 2 (High Ethical Promotion, Low Ethical Promotion) × 2 (High Ethical Demonstration, Low Ethical Demonstration) factorial design. After giving their consent, participants were presented with a workplace scenario asking them to imagine themselves as an employee who observed their manager demonstrating either normatively ethical or unethical behavior while interacting with a customer. They then observed their manager either promoting ethical behavior (accompanied by a request to be ethical at work) or promoting selling behavior (accompanied by a request to set up sale displays) during a meeting. Participants were randomly assigned to one of these four possible conditions (see the appendix for the full manipulations).

Following the scenario and manipulations, participants were asked to report their level of psychological reactance experienced with respect to their manager and their likelihood of engaging in an ethically questionable behavior described in the scenario. We chose to have the manager in both the high and low ethical promotion scenarios make a request of the employee in an attempt to rule out the possibility that employees experience similar levels of psychological reactance in response to all manager requests. To conclude, participants were asked manipulation-check and demographic questions.

Measures

Psychological Reactance

In study 1, we used a measure of psychological reactance that assessed reactance at work in general. In this study, we aimed to capture a more localized experience of psychological reactance—specific to a leader’s particular comments (i.e., promoting ethical or sales-related behavior). We used a 7-item, state-based measure of psychological reactance used in previous research that was focused on the individual who was causing the reactance (Jonas, Graupmann, Kayser, Zanna, Traut-Mattausch, & Frey, 2009). We adapted these items so that they clearly assessed participants’ psychological reactance toward their manager. Participants responded to the following items:

1. 1) How much do you feel your manager’s leadership style restricts your freedom of choice?

2. 2) How much pressure does your manager’s leadership style cause you to feel?

3. 3) How much does your manager’s leadership style bother you?

4. 4) How irritated do you feel about your manager’s leadership style?

5. 5) How much do you feel your manager’s leadership style is reasonable? [reverse coded]

6. 6) How legitimate is your manager’s leadership style? [reverse coded]

7. 7) How much would you rather decide by yourself how to behave at work?

All items were measured on a 7-point scale ranging from 1 (not at all) to 7 (very much). Consistent with previous work using these items (e.g., Jonas et al., 2009), the scale demonstrated good reliability (α = .90), and so we averaged the items to form a single measure.

Deviant Behavior

The scenario suggested that participants may have the opportunity to steal company products or property. Thus, as a measure of deviant behavior, participants were asked a single question: “If the manager didn’t know, how likely would you be to take some of the used products described in the scenario and keep them for yourself?” Participants answered this question on a 7-point scale ranging from 1 (extremely unlikely) to 7 (extremely likely).

Additional Measures

For use as a manipulation check for the ethical promotion condition, participants indicated their agreement with two statements: “Your manager explains what is expected from employees in terms of behaving with integrity” and “Your manager clarifies integrity guidelines.” For use as a manipulation check for the ethical demonstration condition, participants indicated their agreement with three statements: “Your manager can be trusted,” “Your manager will keep promises,” and “Your manager will keep commitments.” All responses were given on a 7-point scale ranging from 1 (strongly disagree) to 7 (strongly agree), and they were averaged together to create measures of the effectiveness of the ethical promotion (α = .95) and ethical demonstration (α = .92) manipulations. In addition to these manipulation-check questions, participants answered an attention-check question and demographic questions.

Results and Discussion

A one-way analysis of variance (ANOVA) was conducted for each pair of experimental conditions (i.e., high ethical promotion vs. low ethical promotion; high ethical demonstration vs. low ethical demonstration) on its respective manipulation-check measure. There was a significant difference in the predicted direction between those in the low and high ethical promotion conditions, F(1, 204) = 53.02, p < .001, η² = .46, as well as in the low and high ethical demonstration conditions, F(1, 204) = 320.07, p < .001, η² = .78. These results increased our confidence that our manipulations had their intended effect.

We conducted a moderated-mediation analysis using model 7 in Hayes’s (2017) PROCESS macro (version 4) for SPSS version 28 to test hypotheses 1a and 2. As in study 1, we examined the interaction between ethical promotion and ethical demonstration on workplace deviance, through psychological reactance as a mediator. Overall, there was a significant interaction effect on psychological reactance, b = −0.62, SE = 0.27, p = .02, 95% CI [−1.17, −0.08]. As can be seen in Figure 3, when ethical demonstration was low, ethical promotion increased psychological reactance, supporting hypothesis 1a. Supporting hypothesis 2, the conditional indirect effect of ethical promotion on deviant behavior—through psychological reactance—was significant when there was low ethical demonstration, b = 0.08, SE = 0.06, 95% CI [0.01, 0.27], but not when the leader engaged in high ethical demonstration, b = −0.05, SE = 0.05, 95% CI [−0.17, 0.02]. Overall, manipulating perceptions of ethical promotion and ethical demonstration, study 2 not only replicates the findings from study 1 but also provides better causal evidence for our prediction that leader moral hypocrisy can trigger psychological reactance among followers—and this reactance can have downstream consequences on employee deviance.

Figure 3: Interactive Effect of Leader Ethical Promotion and Demonstration on Follower Psychological Reactance (Study 2)

Note. Error bars represent standard errors around the mean.

STUDY 3

In study 3, we examined whether alternative mechanisms besides psychological reactance might account for the effect of leader moral hypocrisy on employee deviance. For example, it may be that followers simply do not like or trust hypocritical leaders (e.g., Simons, Friedman, Liu, & McLean-Parks, 2007; Treviño & Brown, 2004), which may result in greater instances of deviance. Moreover, moral hypocrisy has been tied to perceivers’ affect (Valdesolo & DeSteno, 2008), suggesting that leader moral hypocrisy may elicit high levels of negative affect or low levels of positive affect, which then may affect employees’ behavior (Baron, 1993; Lee & Allen, 2002). On the basis of these possibilities, in study 3, we examine four alternatives to psychological reactance that may explain why leader moral hypocrisy can lead to deviant behavior: lower trust, less liking, more negative affect, and less positive affect.

Participants

Two hundred forty undergraduate students from a southern US university participated in this study for partial course credit. We chose to use a university population for the same comprehension and generalizability reasons that we described in study 2. Seven participants were excluded from the sample for not passing an attention check, leaving a final sample of 233 (126 women). Participants had an average age of 19.9 years (SD = 1.30), and 33 percent reported being employed. Seventy-nine percent of participants identified as white/Caucasian, 0.48 percent as Black/African American, 14.63 percent as Asian, 3.90 percent as Hispanic, and 1.96 percent as a race or ethnicity other than those listed.

Study Design and Procedures

Participants were randomly assigned to one of three conditions: ethical leadership, leader moral hypocrisy, and neutral leadership. We included the neutral leadership condition to serve as a truer control condition. Using the same vignette as in study 2, participants in the full ethical leadership condition read the scenario in which their store manager both demonstrated ethical behavior and promoted such behavior in the employee meeting. Those in the leader moral hypocrisy condition read the scenario in which their manager did not demonstrate ethical behavior but later promoted ethics. Those in the neutral leadership condition read that the manager simply relayed facts rather than demonstrating ethics and did not promote ethics during the meeting.

After participants read the scenario, they answered questions about their psychological reactance, their trust in the manager, how much they liked their manager, and their own positive and negative affect after the employee meeting. Following these questions, participants were asked to rate their likelihood of engaging in the ethically questionable behavior described in the scenario. To conclude, participants answered manipulation-check and demographic questions.

Measures

Psychological Reactance

We used the same scale as in study 2 to measure psychological reactance (α = .89).

Trust

We measured trust using three items developed by Mooijman, Van Dijk, Van Dijk, and Ellemers (2019), such as “This manager can be fully trusted.” Participants responded to the items using a 7-point scale ranging from 1 (strongly agree) to 7 (strongly disagree) (α = .94).

Liking

Liking was measured using three items, including “I would enjoy spending more time with this manager” (Huang, Yeomans, Brooks, Minson, & Gino, 2017). Participants responded using a 7-point scale ranging from 1 (strongly agree) to 7 (strongly disagree) (α = .94).

Positive and Negative Affect

We used the Positive and Negative Affect Scale (PANAS), developed by Watson, Clark, and Tellegen (1988), to measure both positive affect (PA; α = .92) and negative affect (NA; α = .87). The scale comprises twenty items (ten items relating to PA and ten items relating to NA). Participants provided their responses using a 5-point scale ranging from 1 (not at all) to 5 (extremely).

Deviant Behavior

We used the same measure of deviant behavior as in study 2, in which participants indicated the likelihood that they would steal used products from the company.

Manipulation Checks

We used the same measures from study 2 as our manipulation-check questions for ethical promotion (α = .90) and ethical demonstration (α = .93).

Results and Discussion

First, we tested our manipulations by conducting a one-way ANOVA with planned contrasts using comparisons for all three conditions to each other on the manipulation-check measures. We began by looking at ethical promotion in each condition. We coded leader moral hypocrisy as 0 and the comparison conditions as 1 in all comparisons. Those in the neutral leadership condition rated their manager as providing less ethical promotion than those in the leader moral hypocrisy condition, t(202) = −5.18, p < .001. Those in the ethical leadership condition rated their manager as providing more ethical promotion than both those in the leader moral hypocrisy condition, t(202) = 4.88, p < .001, and those in the neutral leadership condition, t(202) = 9.98, p < .001. Next, we looked at ethical demonstration in each condition. As expected, those in both the neutral leadership and ethical leadership conditions rated their manager as providing more ethical demonstration than those in the leader moral hypocrisy condition, t(202) = 4.79, p < .001, and t(202) = 10.97, p < .001, respectively. Those in the ethical leadership condition reported higher levels of ethical demonstration than those in the neutral leadership condition, t(202) = 6.09, p < .001. Together, these results offer support that our manipulations worked as intended.

In this study, we sought to replicate the mediation effect found in studies 1 and 2 and rule out potential alternative mediators. We conducted mediation analyses using model 4 of Hayes’s (2017) PROCESS macro (version 4) for SPSS version 28. We created dummy codes for each condition so that we could compare the effect of leader moral hypocrisy against both neutral leadership and ethical leadership. In addition to psychological reactance, we included trust, liking, positive affect, and negative affect in the model as mediators.

We first examined the differences between ethical leadership (coded as 0) and leader moral hypocrisy (coded as 1). Leader moral hypocrisy led to more psychological reactance, b = 1.76, SE = 0.17, p < .001, 95% CI [1.42, 2.10], supporting hypothesis 1a (see Table 3). There were also significant differences between leader moral hypocrisy and ethical leadership in relation to liking, trust, and negative affect. Positive affect, however, showed no difference across leadership type (see Table 3). Despite leader moral hypocrisy affecting several reactions from employees, only psychological reactance seemed to drive employee deviant behavior, b = 0.36, SE = 0.16, p = .03, 95% CI [0.03, 0.68]; liking, trust, positive affect, and negative affect had no significant effect on deviant behavior when included in the model with psychological reactance (see Figure 4). Furthermore, the indirect effect of leader moral hypocrisy on deviant behavior was statistically significant only through psychological reactance, b = 0.63, SE = 0.29, 95% CI [0.07, 1.21]. As can be seen in Table 3, the confidence intervals for liking, trust, positive affect, and negative affect all crossed zero, suggesting that they did not impact the relationship between leader moral hypocrisy and unethical behavior. Together, these results support hypothesis 2.

Table 3: Summary of Mediation Results Comparing Leader Moral Hypocrisy to Ethical Leadership and Neutral Leadership (Study 3)

Note. Leader moral hypocrisy is dummy coded (i.e., 1 = leader moral hypocrisy condition, 0 = comparison condition). Standard errors are in parentheses; 95 percent confidence intervals are in square brackets. Indirect effect 95 percent confidence intervals are based on 10,000 bootstrap samples. PR = psychological reactance, PA = positive affect, NA = negative affect.

* p < .05.

** p < .001.

Figure 4: Mediation Model Comparing the Effect of Leader Moral Hypocrisy to Ethical Leadership on Deviant Behavior (Study 3)

Note. The leader moral hypocrisy dummy variable is coded as follows: 0 = ethical leadership, 1 = leader moral hypocrisy. Values in square brackets represent the total effect once mediators are included in the model. Other brackets represent the standard error of the effect. ^p < .10. *p < .05. ***p < .001.

The pattern of results was largely the same when comparing leader moral hypocrisy (coded as 1) to neutral leadership (coded as 0). Again, leader moral hypocrisy led to more psychological reactance, b = 0.79, SE = 0.17, p < .001, 95% CI [0.45, 1.13]. There were also significant differences between leader moral hypocrisy and neutral leadership in relation to liking and trust. However, both positive and negative affects showed no difference across leadership type (see Table 3). Once again, only the indirect effect of leader moral hypocrisy on deviant behavior through psychological reactance was statistically significant, b = 0.28, SE = 0.15, 95% CI [0.03, 0.61]. The confidence intervals for liking, trust, positive affect, and negative affect all crossed zero (see Table 3), suggesting that they did not impact the relationship between leader moral hypocrisy and unethical behavior. Together, these results offer additional support for hypothesis 2.

It is perhaps unsurprising that leader moral hypocrisy influenced liking, trust, and negative affect (when compared to ethical leadership) and liking and trust (when compared to neutral leadership). However, it is interesting that psychological reactance seemed to better explain the effect of leader moral hypocrisy on deviant behavior when all the potential mediators were included in the model together. These results support hypothesis 2 and strengthen our argument that psychological reactance is a key factor in a person’s deviant behavior in response to leader moral hypocrisy. In study 4, we therefore delved deeper into our examination of psychological reactance to explore a potential mechanism behind why it is triggered by leader moral hypocrisy.

STUDY 4

To more fully understand why leader moral hypocrisy can lead to psychological reactance, in this study, we examined the role of followers’ perceptions of their leaders’ use of power. We employed a similar experimental paradigm as used in studies 2 and 3, but in this study, we randomly assigned participants to only one of two conditions: leader moral hypocrisy or ethical leadership. ³

Participants

Three hundred one participants (63.46% women) were recruited from Prolific, an opt-in, online panel of individuals willing to participate in academic surveys for payment. Participants had an average age of 41.74 years (SD = 14.56), and 74.09 percent reported being employed. In our sample, 83.72 percent of participants identified as white/Caucasian, 4.32 percent as Black/African American, 5.32 percent as Asian, and 6.64 percent as a race or ethnicity other than those listed.

Study Design and Procedure

After agreeing to participate in the study, participants were randomly assigned to read the leader moral hypocrisy or ethical leadership scenario described in study 3. Following the scenario, participants were asked questions about their manager’s use of power and their level of psychological reactance experienced with respect to their manager’s request. Participants concluded by answering manipulation-check, attention-check, and demographic questions.

Measures

Perceived Use of Power

Participants were asked to rate their perceptions of their manager after the employee meeting. They rated their agreement on a 7-point scale ranging from 1 (strongly disagree) to 7 (strongly agree) with three statements: “My manager relies on power to influence me and my co-workers,” “My manager uses power to get performance results from me and my co-workers,” and “My manager uses coercive tactics when trying to get me and my co-workers on board with an idea.” These items demonstrated good reliability (α = .88) and were averaged together to form a single measure.

Psychological Reactance

We used the same measure of psychological reactance as in studies 2 and 3 (α = .92).

Manipulation Checks

We used the same manipulation checks for ethical promotion and demonstration as in studies 2 and 3. Both the ethical promotion (α = .89) and ethical demonstration (α = .95) manipulation-check questions were averaged to create measures of the effectiveness of each manipulation.

Results and Discussion

As preregistered, eight participants were removed from the data analysis for failing an attention check or not finishing the study. We conducted a t test to demonstrate that participants in both conditions (combined) rated their managers’ ethical promotion as significantly higher than the midpoint of the scale, t(292) = 22.92, p < .001 (because both conditions were intended to display the promotion of ethics by the manager). We then conducted a one-way ANOVA, which demonstrated that those in the leader moral hypocrisy condition rated their manager as demonstrating less ethical behavior than those in the ethical leader condition, F(1, 292) = 297.14, p < .001, η² = .51. These results increased our confidence that our manipulations had their intended effect.

We conducted a mediation analysis using model 4 of Hayes’s (2017) PROCESS macro (version 4) for SPSS version 28 to test the mediating effect of the perceived use of power on the relationship between leader moral hypocrisy and psychological reactance. We coded the conditions so that 0 represented ethical leadership and 1 represented leader moral hypocrisy in our data set. First, we again found that the effect of leader moral hypocrisy on psychological reactance was positive and significant, b = 1.88, SE = 0.12, p < .01, 95% CI [1.63, 2.12], supporting hypothesis 1a. Moreover, leader moral hypocrisy increased followers’ perceptions of their leader’s use of power, b = 1.59, SE = 0.14, p < .001, 95% CI [1.33, 1.86], and the leader’s perceived use of power significantly affected psychological reactance, b = 0.48, SE = 0.05, p < .001, 95% CI [0.39, 0.57]. Supporting hypothesis 1b, we found a significant indirect effect of leader moral hypocrisy on follower psychological reactance through the leader’s perceived use of power, b = 0.77, SE = 0.09, 95% CI [0.59, 0.95]. However, when perceived use of power was included in the model, leader moral hypocrisy still had a significant (albeit reduced) effect on psychological reactance, b = 1.11, SE = 0.13, p < .001, 95% CI [0.86, 1.36], suggesting that the perceived use of power partially mediates the relationship. Thus the results of this study provide evidence that perceptions of a leader’s use of power help to explain, at least partly, why leader moral hypocrisy can trigger psychological reactance in followers.

GENERAL DISCUSSION

There is little doubt that ethical leadership matters at work. Previous research has focused on the behaviors that result from ethical leadership, demonstrating that it can increase employee performance and organizational citizenship behavior (Piccolo, Greenbaum, Den Hartog, & Folger, 2010). However, we add to mounting evidence that there is also little doubt that leader moral hypocrisy matters at work. We demonstrate that leaders who promote but do not demonstrate ethics (i.e., leader moral hypocrisy) are seen as exercising more power, which can result in followers experiencing psychological reactance. Such reactance can then lead followers to engage in increased unethical behavior. In our first three studies, we demonstrated the role of psychological reactance in the link between leader moral hypocrisy and follower unethical behavior. Study 1 used a sample of employee–manager dyads to show that employees reported higher levels of psychological reactance at work when managers were rated higher on ethical promotion but lower on ethical demonstration (i.e., they demonstrated leader moral hypocrisy). Employees who reported higher levels of psychological reactance were then more likely to exhibit deviant behavior, as reported by their managers. We found similar patterns in studies 2 and 3 after experimentally manipulating followers’ perceptions of a leader’s moral hypocrisy, which increased our confidence in making causal inferences. Notably, in study 3, we ruled out several alternative explanations for why employees might engage in unethical behavior. Finally, study 4 offered further insight into what may drive follower psychological reactance, finding support for the explanation that perceptions of a leader’s use of power help to explain this link.

Theoretical Implications

Our findings contribute to a better understanding of the consequences of leader moral hypocrisy by introducing psychological reactance as an important psychological outcome. Early work on ethical leadership suggested that without ethical demonstration, ethical promotion would be seen as hypocritical (Treviño & Brown, 2004; Treviño et al., 2000). However, the negative effects of such hypocrisy were described primarily from the perspective of social learning theory, whereby a hypocritical leader’s attempts to promote ethics would fall flat and be dismissed as window dressing, with followers learning vicariously from their leader’s behavior and withdrawing from the situation (Treviño et al., 2000). By further exploring the construct of psychological reactance, we show that followers may not merely “dismiss” the ethical promotion of a hypocritical leader but be motivated to actively work against it. We provide evidence suggesting that at the individual level, ethical promotion alone can lead to increased psychological reactance among followers.

Relatedly, our research offers further insight into the construct of psychological reactance—a construct that is not well explored in management research but likely highly relevant. In addition to reviewing previous research on this construct, we developed theory on why leader moral hypocrisy may be especially likely to result in follower psychological reactance. More specifically, we introduced the notion that by only promoting—but not demonstrating—ethics to followers, leaders were seen as exercising more coercive power, placing undue pressure on followers to engage in certain behaviors and causing them to experience psychological reactance. However, our research suggests that beyond the specific context of leader moral hypocrisy, reactance may also be a negative side effect that emerges in a broad range of organizational contexts, particularly those related to situations involving social influence and other forms of leadership. Given the natural hierarchies inherent in organizational life, our findings suggest that a good starting point for finding psychological reactance in organizations might be any place where the legitimacy or moral authority of the hierarchy is in question, especially as it relates to moral situations.

Our work also extends what we know about leadership and ethics in general by looking at how the two components of ethical leadership relate to one another and affect followers’ attitudes and behaviors. Although Mayer et al. (2012) highlighted the need to differentiate between the moral person and the moral manager, most prior research has assumed that ethical leadership is a single, unified construct and thus turned its focus to comparisons with other forms of leadership (Avolio & Gardner, 2005; Hassan, Mahsud, Yukl, & Prussia, 2013; Luthans & Avolio, 2003). As a result, there is little understanding regarding how the two components of ethical leadership might uniquely influence follower behavior. Across our four studies, we find that ethical promotion on its own can undermine leaders’ attempts to encourage followers to refrain from engaging in unethical behaviors. Though our findings demonstrate an important pattern in relation to unethical behavior, we believe that further insight can be gained by exploring these constructs independently in relation to other common outcomes of ethical leadership, including organizational citizenship behaviors and overall performance.

Finally, although we focused on the negative implications of leaders’ hypocritical displays of ethical promotion, our research also offers insight into the important role that ethical demonstration plays in discouraging negative behaviors. Research related to ethical demonstration has suggested that the personal demonstration of promoted behaviors may well serve, as Simons et al.’s (2012) review of behavioral integrity suggests, as a moderator for the effectiveness of many or even most leader initiatives. Many leadership efforts either succeed or fail contingent upon the leader’s personal credibility. However, this potentially critical contingency has been largely unstudied (for an exception, see Dineen, Lewicki, & Tomlinson, 2006). Our work shows that ethical demonstration may be the linchpin to ensure ethical credibility and may further determine a leader’s success in changing ethical decisions among followers. Indeed, it is crucial that leaders “walk the talk” to reduce the likelihood that followers will experience psychological reactance. Doing so can not only help leaders avoid unintentionally encouraging undesired behaviors but also help them improve employee attitudes and encourage more desirable behavior among followers (Davis & Rothstein, 2006).

Limitations and Future Directions

As with any set of studies, ours are not without limitations. First, although study 1 provided an opportunity to examine the effects of ethical promotion and ethical demonstration in an organizational context, the setting and study design provided some restrictions as well. One restriction was that study 1 was correlational in design, limiting the causal inferences that could be made. Additionally, although we aimed to avoid common method bias by collecting responses from both followers and their leaders, we recognize that social desirability biases may have been at play when leaders rated their own levels of ethical guidance. We designed studies 2 and 3 to address this and other shortcomings of study 1. These latter studies, however, had limitations that study 1 did not. For example, the experiments were based on a hypothetical scenario. Real situations, especially involving morality and ethics, may trigger affective reactions that are simply not present when reading a scenario. Another limitation of our studies was that only one (study 4) examined perceptions of the leader’s power as an explanation for why leader moral hypocrisy leads to psychological reactance.

These limitations and the findings of our studies point to several important avenues for future research. First, as mentioned earlier, we believe that future research should further explore the role of psychological reactance in the workplace. For example, under what conditions do perceived restrictions of freedom at work not lead to psychological reactance? Perhaps it is the source of the restriction that matters (e.g., leader directives, monitoring systems), or maybe the perceived consequences associated with noncompliance (e.g., threats of punishment).

Second, given that a leader’s perceived use of power only partially mediated the relationship between leader moral hypocrisy and follower psychological reactance, future research should delve more deeply into the mechanisms that spark such reactance. What other factors (if not trust, liking, and affect, which we explored in study 3) help explain the relationships we observed, and what moderating factors might reduce or exacerbate the negative effects of leader moral hypocrisy on psychological reactance?

Finally, although our research suggests that psychological reactance may be particularly likely in situations pertaining to hypocritical leadership, many daily work routines involve employees receiving some form of direction from a leader that potentially restricts their autonomy. We know very little about when such directives result in psychological reactance. Our work suggests that moral directives coming from someone who has forfeited their moral authority may be particularly likely to provoke psychological reactance, but there are likely other contexts and types of leadership directives that lead to psychological reactance, whether they be moral or not. A key moderating influence likely relates to the severity of the perceived restriction of freedom (i.e., how hard or how much someone or something is pushing). Future research might fruitfully examine different situations and types of influence attempts at work (in terms of the severity of the restrictions they place on employee autonomy) and their various effects on employee psychological reactance. Overall, organizational research is ripe for a focused examination of the antecedents and consequences of psychological reactance in the workplace.

CONCLUSION

Ethical leadership has been shown to be an effective solution for leaders who want more ethical, and less unethical, behavior from their employees. However, the possibility that the two components of ethical leadership (i.e., ethical promotion and ethical demonstration) are not always enacted simultaneously is often ignored—as are the negative consequences that can result. By offering insight into a psychological consequence among followers that stems from leader moral hypocrisy, namely, psychological reactance, our work offers researchers a launching pad from which to better understand (and, it is hoped, prevent) the role of both leader moral hypocrisy and psychological reactance in motivating employees to engage in deviant and unethical behaviors at work.