Introduction

The corporate scandals that have come to the general public’s attention in the past years (e.g., Enron and Fannie Mae’s accounting practices, HP and Deutsch Bank’s spying accusations) have again highlighted the importance of the ethical conduct of those in influential positions. Much has been written about the normative or philosophical view of ethical leadership (see Brown and Treviño 2006 and Eisenbeiss 2012 for a review), and the ethical dimension is present in the contemporary models of leadership (e.g., idealized influence dimension of transformational leadership—Bass and Avolio 1994). Yet, and while organizations have started to express the importance of having leaders that behave ethically (Stouten et al. 2012), ethical scandals are still commonplace.

Although it has been in the mind of researchers, practitioners, and the general public for a long time, only recently the construct (and corresponding measure) of ethical leadership was developed (Brown et al. 2005). These authors have defined ethical leadership as “the demonstration of normatively appropriate conduct through personal actions and interpersonal relationships, and the promotion of such conduct to followers through two-way communication, reinforcement, and decision-making” (p. 120). This definition highlights three key characteristics: (a) the leader is a legitimate and credible role model; (b) the leader not only explicitly emphasizes the importance of ethics, but also reinforces ethical behaviors (while disciplines those that do not comply with the standards); and (c) the leader evaluates the ethical implications of his/her decisions.

Since Brown et al.’s (2005) seminal paper, researchers started to examine how ethical leadership affects employees’ conduct, i.e., the processes through which its influence takes place. Researchers have highlighted multiple mechanisms, including the work environment (poor working conditions: Stouten et al. 2010; ethical climate and culture: Neubert et al. 2009; Mayer et al. 2010; Schaubroeck et al. 2012; Shin 2012), job characteristics (workload: Stouten et al. 2010; task significance and autonomy: Piccolo et al. 2010) and employees’ (individually or as a group) trust in leader (Newman et al. 2013) work engagement (Den Hartog and Belschak 2012), well-being (Kalshoven and Boon 2012), voice (Avey et al. 2012; Walumbwa et al. 2012), psychological ownership (Avey et al. 2012), safety (Walumbwa and Schaubroeck 2009), and conscientiousness (Walumbwa et al. 2012).

Despite this growing evidence concerning the mechanisms linking ethical leadership and employee outcomes, authors still highlight the need for further research in this area (e.g., Mayer et al. 2012; Newman et al. 2013). Moreover, the boundary conditions that limit or enhance the effectiveness of ethical leadership have received much less attention and usually emphasize either the role played by the work environment, such as organizational politics (Kacmar et al. 2011) and human resource management practices (Kalshoven and Boon 2012) or by employees’ characteristics, such as self-esteem (Avey et al. 2011) or gender (Kacmar et al. 2011). Further research on the boundary conditions of ethical leadership is particularly relevant as it may help explain under what conditions ethical leadership becomes more valuable for organizations and influential for employee behavior.

Thus, the present study aims to contribute to the discussion concerning the process of ethical leadership, namely how and under what circumstances it influences employee behavior. First, we extend our knowledge of the process of ethical leadership by examining the role of social exchanges. Social exchange theory (Blau 1964) has been a predominant framework in organizational behavior (Cropanzano and Mitchell 2005), and particularly in the study of leadership (Dirks and Ferrin 2002; Tse et al. 2013). Nonetheless, and while it has been discussed (theoretically) as a key dimension in the relationship between ethical leaders and their subordinates (Brown and Treviño 2006), the social exchange mechanisms through which ethical leaders influence follower behaviors are yet to be examined empirically (ibidem). For example, Brown and Treviño (2006) argue that through a high quality exchange relationship with managers, ethical leadership should decrease the likelihood of engagement in negative behaviors—yet, such mechanisms are not detailed in their propositions. Therefore, we propose and test that ethical leadership contributes to the reduction of employee’s deviant behaviors through an increase in employees’ affective organizational commitment, one of the main indicators of a high quality social exchange relationship with the organization (Colquitt et al. 2013), since the actions of the supervisor are, to some extent, interpreted as actions on behalf of the organization.

Second, we further develop our understanding of the conditions that significantly affect the strength of the impact of ethical leadership. While there is an overall agreement that the demonstration of high moral standards by leaders is an important determinant of employee behavior (see Brown and Treviño 2006 for a review), there is also evidence (although this line of inquiry is still in its infancy) that the impact of such actions varies substantially between individuals (e.g., Avey et al. 2011) or across contexts (Kalshoven and Boon 2012). Given that one of the main misconceptions about ethical leadership is that ethics and effectiveness are not compatible (Brown 2007), visible in the popular adage “nice guys finish last” (ibidem), it is urgent that researchers provide empirical tests of this assumption. In this paper, we suggest that the role ethical leadership plays in the employee–organization relationship (EOR) varies significantly depending on the supervisor’s reputation for performance, such that ethical leadership should be more effective when leaders also demonstrate a high potential for success. High ethical leadership combined with high reputation for performance indicates that the supervisor not only fulfills his/her performance obligations at the workplace, but does so following a set of moral standards. Finally, and based on a social exchange framework, we expect that employees reduce their deviant behaviors as a response to ethical leadership (conditional on supervisor’s reputation for performance), through increased affective commitment to the organization (i.e., moderated mediation effect).

Overall, this paper adds to the literature in two main ways by highlighting: (a) the employee–organizational relationship as a key social exchange mechanism through which ethical behavior by leaders, as agents of the organization, reduces deviant behaviors directed at the organization; and (b) reputation for performance, as an indication that the leader is competent and helpful, as an employee perception that significantly shapes how employees react to ethical leadership.

Ethical Leadership and the EOR

As we previously discussed, researchers have begun to explore the mechanisms linking ethical leadership and employee behaviors. However, and while multiple aspects of the organizational life have been examined as potential mechanisms, the EOR has been largely overlooked, despite its key role for organizational functioning (see Rhoades and Eisenberger 2002 and Meyer et al. 2002 for meta-analyses on perceived organizational support, POS, and organizational commitment, respectively). These two concepts capture both sides of the positive EOR: how employees perceive the treatment received by the organization (POS: Eisenberger et al. 1986) and how employees characterize their relationship with the organization (Meyer and Allen 1991).

The EOR is “an overarching term to describe the relationship between the employee and the organization” (Shore et al. 2004, p. 292) and is drawn on social exchange theory (Blau 1964). Social exchange theory posits that unlike the specific benefits or commodities involved in an economic exchange, the favors or benefits exchanged are indicative of mutual support and investment in the relationship (Aryee et al. 2002). As such, only social exchanges tend to engender feelings of personal obligation, gratitude, and trust; purely economic exchanges do not (Blau 1964).

Underlying the application of social exchange theory to the EOR are the key assumptions that: (a) the employee attributes humanlike characteristics to the organization; and (b) organizational agents pursue the organization’s interests in their relationship with their subordinates (Coyle-Shapiro and Shore 2007). In fact, the personification of the organization is actually strengthened by the legal, moral, and financial responsibility for the actions of its representatives (Eisenberger et al. 1986). Moreover, the role of direct supervisors as agents of the organization is critical since they are the ones responsible for coordinating work, assigning tasks, and evaluating the performance of employees.

The norms governing social exchanges, based on reciprocity, predict that an imbalanced relationship would evolve over time in order to achieve congruence, that is, either mutual positive or negative feelings. Thus, it is of interest to understand whether employees reciprocate ethical leadership, as behaviors enacted by an organizational representative (i.e., supervisors), by stimulating the bond employees develop with their organization, particularly “the emotional attachment to, identification with, and involvement in the organization,” i.e., affective organizational commitment (Meyer and Allen 1991, p. 67). As Cropanzano and Mitchell (2005) highlighted, commitment, together with trust and loyalty, forms one of the basic tenets of social exchanges. The reasoning is that virtuous behavior—based on principles of fairness and integrity—from leaders contributes to the thriving of all members of the community (Neubert et al. 2009) and as such would strengthen the emotional bond between employees and their organization, one of the motivational mechanisms supporting social exchanges in the workplace.

Previous research has showed that employees generalize their views concerning the actions of their supervisors to the organization. In fact, Meyer et al. (2002) argued that among the things the organization can do to enhance employee commitment is to provide strong leadership, since employees view their leaders as agents working on the organization’s behalf. Their meta-analytical findings supported their argument and presented a strong correlation between transformational leadership (which encompasses ethical behaviors) and affective organizational commitment. The few studies that have examined the relationship between ethical leadership and affective organizational commitment (Den Hartog and De Hoogh 2009; Kalshoven et al. 2011; Neubert et al. 2009) provide strong empirical support to this claim, since, and despite using different measures and conceptualizations of ethical leadership, this relationship is fairly stable and strong. Thus,

Hypothesis 1

Ethical leadership is positively related to affective commitment to the organization.

The Moderating Role of Reputation for Performance

Ethical leadership, as an influence process between leaders and subordinates, is influenced by the general character the latter attribute to the leader. One key feature of the character of leaders employees try to assess is personal reputation. As Hochwarter et al. (2007) put it, examining influence processes without considering how they relate to different reputations provides a fairly incomplete perspective. Personal reputation concerns the positive image one tries to develop based on how other individuals perceive them as performing their jobs competently and being helpful in the workplace (Zinko et al. 2012). It is defined as a “complex combination of salient characteristics and accomplishments, demonstrated behavior, and intended images presented over some period of time” (Ferris et al. 2003, p. 213).

One of the arguments put forth to explain why personal reputation acts as a boundary condition for influence processes is that it serves an important signaling function (Posner 1997), carrying information about an individual’s intentions. For example, Laird et al. (2009) found that an increase in felt accountability in the workplace presented quite different results, depending on the person’s reputation. Similarly, Hochwarter et al. (2007) found that the relationship between political behavior and several outcomes (uncertainty, exhaustion, and performance) varied significantly depending on the levels of personal reputation. In both studies, reputation determined whether political behavior and accountability presented positive (when reputation was high) or negative (when reputation was low) outcomes.

As one can develop reputations for a myriad of things (e.g., a person that has a reputation for being highly creative does not necessarily have a reputation to deliver outputs on time), in the present study we are particularly interested in the reputation concerning the supervisor’s ability to perform one of the high-order dimensions that compose personal reputation in the context of organizations (Zinko et al. 2012). This dimension of personal reputation concerns whether employees see the supervisor as someone that “gets things done,” which is crucial since ethical leaders are also concerned with how people perform their jobs (Stouten et al. 2010). Thus, while consistent ethical supervisor behaviors should always contribute positively to the EOR, we argue that employees put their assessments of ethical leader behavior in perspective and look for additional information concerning the leader’s reputation for performance, as they look for additional signals (Posner 1997) to try to make sense of the implications of his/her ethicality.

A second argument for the examination of reputation for performance in our model concerns its role in the assessment of the leader’s trustworthiness, a key input for the development of the EOR. The characteristics and actions of individuals make others trust them more or less (Mayer et al. 1995). For example, ethical leadership positively contributes to trust in the leader, mainly because it signals that the leader has integrity, a key dimension of trustworthiness (Newman et al. 2013). Another important characteristic of the trustee is competence (or ability), the expertise, knowledge, and skills that enable a party to have influence within some specific domain (Mayer et al. 1995). Reputation for performance is built around demonstrations of competence as a leader, which influences not only trust but also affective commitment to the organization (Colquitt et al. 2007), two of the principal mechanisms in social exchanges (Colquitt et al. 2013).

Finally, the examination of reputation for performance provides an important contribution to one of the key misconceptions about ethical leadership identified by Brown (2007), that ethics and effectiveness are not compatible “in the rough and tumble business world” (p. 149). As Brown (2007) highlighted, the preliminary data concerning how ethical leadership is related to positive outcomes are compelling. Yet, we are not aware of any studies that examine the intertwinement between assessments of leader ethics and effectiveness and how it influences employee behavior. In light of these arguments, we believe the examination of the interplay between ethics and effectiveness is central for a deeper understanding of the ethical leadership process. Therefore, we suggest that a positive reputation for performance will enhance the effectiveness of ethical leadership in the development of an emotional bond between employees and the organization, by (a) signaling that the leader is someone that delivers but that does not follow the maxim “by all means necessary” and (b) showing that the leader is trustworthy. A negative reputation for performance should lessen the effectiveness of ethical leadership, as it would signal the leader is someone with good intentions but unable to provide what the organization requires, and, therefore, untrustworthy, which should also affect employees’ organizational life as the organization may limit the amount of resources provided to the supervisor.

Hypothesis 2

Reputation for performance moderates the relationship between ethical leadership and affective commitment to the organization, such that when reputation for performance is high the positive relationship is stronger than when reputation for performance is low.

Carryover Effects on Organizational Deviance

Researchers have been particularly concerned with the relationship between ethical leadership and employee behaviors, particularly performance (e.g., Kacmar et al. 2011; Shin 2012; Walumbwa et al. 2012) and deviance (e.g., Avey et al. 2011; Den Hartog and Belschak 2012; Mayer et al. 2009). As we previously discussed, social exchange theory (Blau 1964) has provided the conceptual underpinning of research on work attitudes and behaviors (Settoon et al. 1996; Wayne et al. 1997), and affective commitment to the organization is one of its key mechanisms. In general, positive treatment received from the supervisor makes employees develop a felt obligation to reciprocate those positive actions, mostly through an increase in affective commitment to the organization (Eisenberger et al. 2010). That is, subordinates feel indebted and obliged to repay organizations and its representatives (i.e., supervisors) in kind over time (Walumbwa et al. 2009).

These behaviors (i.e., performance and deviance) represent two forms of reciprocating the treatment received by the organization and its agents that have an important impact on organizational functioning. It is particularly important to further analyze deviant behaviors directed at the organization (e.g., intentionally working slower, come in late without permission, lie about the number of hours worked) for three reasons: (a) workplace deviance is an expensive and pervasive problem for organizations that happens rather frequently (Bennett and Robinson 2000); (b) it represents how much employees engage in “voluntary behaviors that violate significant organizational norms and in so doing threatens the well-being of an organization its members or both” (Robinson and Bennett 1995, p. 556), which is an important behavioral indicator of the negative employee view of the EOR, since it is specifically directed at the organization and carries the implications described in the previous point; and (c) it results from a deliberate desire to violate organizational norms derived from basic moral standards (Feldman 1984), including those upheld by organizational agents such as supervisors.

Moreover, we could only find a handful of studies that examine the relationship between ethical leadership and deviant behaviors, and only two of those studies examined potential pathways and moderators of such relationship. Den Hartog and Belschak (2012) found that ethical leadership reduced deviant (or counterproductive) behaviors through an increase in work engagement, but that increase varied depending on the levels of leader Machiavellianism. In another study, Avey et al. (2011) found that ethical leadership decreased the engagement in deviant behaviors only when subordinate’s self-esteem was low.

Thus, we propose that, based on social exchange theory (Blau 1964) and the norm of reciprocity (Gouldner 1960), ethical leadership should affect subordinates’ behavior through an increase in affective commitment to the organization, conditional on the supervisor’s personal reputation for performance. As ethical leadership increases, employees become more committed to the organization (since supervisors personify, to some extent, the organization and treatment received by the supervisor is also attributed to the organization), particularly when the supervisor has a high reputation for performance (as an indication that the leader is also known for “delivering the goods”), which should in turn reflect itself, as a result of the norm of reciprocity, in the reduction of voluntary negative behaviors directed at the organization.

Hypothesis 3

Affective organizational commitment is negatively related to organizational deviance.

Hypothesis 4

The indirect effect of ethical leadership on organizational deviance through affective commitment to the organization will be stronger when reputation for performance is high.

Overall, our conceptual model describes a moderated mediation, i.e., a conditional indirect effect model, where the magnitude of an indirect effect at particular values of the moderator is examined (Preacher et al. 2007). Specifically, we examine if the relationship between ethical leadership and organizational deviance via affective organizational commitment is conditional on supervisor’s reputation for performance (see Figs. 1, 2).

Fig. 1
figure 1

Conceptual model

Full size image
Fig. 2
figure 2

Plot of the interaction effect between ethical leadership and reputation for performance

Full size image

Methods

Sample and Procedures

Initially, 25 organizations were identified and contacted to participate in a larger study on Ethics and Social Responsibility. Organizations were identified by either having a strong commitment toward ethics and social responsibility as highlighted on their website or if they had appeared on the media about their ethics and social responsibility program. Corresponding organizational representatives (top management or/and HR managers or/and Public Relations office) were then contacted asking to participate in the study. Eighteen organizations (72 % response rate) agreed to participate by having a sampled supervisor and corresponding subordinate to fill out a survey. We only provided the survey if both, employee and respective supervisor, were willing to participate in the study. In our first contact, we requested 364 subordinate and supervisor participants to participate, however, only 273 agreed to take the survey, representing a 75 % response rate. This high response rate can be explained by the fact that surveys were handed out personally. We had to exclude 49 dyads, because they did not complete the survey or did not have corresponding supervisors/subordinates surveys completed. Our final sample consisted of 224 subordinate–supervisor dyads.

The 18 sampled organizations were from a variety of industry, including: Tourism (3 %), Services (4 %), Health (11 %), Consulting (13 %), Transportation (13 %), Energy (20 %), and Financial Services (31 %). Out of the supervisor sample, 54 % were male with mean age of 44, ranging from 25 to 62 years old. Thirty-six percent of them have worked in the organization for more than 20 years, 32 % between 10 and 20 years, and 27 % between 5 and 10 years. Forty-two percent of the supervisor sample had at least a Master’s degree with 32 % having a Bachelor’s degree. Subordinate population was 56 % female with 40 % having a Bachelor’s degree and 26 % having at least a Master’s degree. Average age for subordinates participants were 35 ranging from 18 to 63 years old. In terms of tenure, 31 % of them have worked in the organizations between 1 and 5 years, 21 % between 5 and 10 years, and 20 % between 10 and 20 years. Supervisor sample rated their respective subordinate organizational deviance, while subordinate sample rated their leaders on ethical leadership and reputation for performance and auto-reported their organizational commitment.

Measures

Control Variables

Previous research has found that age, tenure, gender, and education are sometimes related to affective commitment to the organization and organizational deviance (Berry et al. 2007; Meyer et al. 2002). Therefore, our set of analysis included these demographic variables as controls.

Ethical Leadership

Ethical leadership was rated on 10 items developed by Brown et al. (2005). Sample items include “my supervisor has the best interests of employees in mind’’ and ‘‘my supervisor sets an example of how to do things the right way in terms of ethics.’’ Cronbach’s α was 0.88.

Reputation for Performance

Reputation for performance was rated on six items taken from the Personal Reputation scale developed by Hochwarter et al. (2007). This set of items specifically measures one’s reputation for performance. Sample items include ‘‘Others regard my supervisor as someone who gets things done’’ and ‘‘People know my supervisor produce only high quality results.’’ Cronbach’s α was 0.74.

Affective Organizational Commitment

Affective organizational commitment was rated using the 6 items developed by Meyer et al. (1993). Sample questions include ‘‘I would be very happy to spend the rest of my life in this organization’’ and ‘‘I do not feel emotionally attached to this organization’’ (reversed). Cronbach’s α was 0.75.

Organizational Deviance

Organizational deviance was rated by supervisors on 9 items (Aquino et al. 1999). Sample questions include ‘‘This subordinate took undeserved breaks to avoid work’’ and ‘‘this subordinate left work early without permission.’’ Cronbach’s α was 0.80.

All scales used 5-point Likert-type scales, anchored in 1—Completely disagree and 5—Completely agree.

Results

Descriptive statistics, reliabilities, and correlations are presented in Table 1. The strong correlation between ethical leadership and reputation for performance is noteworthy. Despite being theoretically distinct, in light of the trustworthiness literature such a correlation is not surprising. When developing their model of trustworthiness, Mayer et al. (1995) emphasized that although each factor captures some unique elements of trustworthiness, the fact that they are separable does not mean they are unrelated. If we examine the pattern of relationships between integrity and competence, two of the underlying factors of trustworthiness that are closely related to ethical leadership and reputation for performance, respectively, we find similar results (Colquitt and Rodell 2011; Colquitt et al. 2007; Mayer and Davis 1999). In these studies, the zero-order correlation between integrity and competence ranged from 0.68 (in the meta-analysis conducted by Colquitt et al. 2007) to 0.78, but the factor analyses always supported the distinction between the dimensions. Our results point in the same direction as the trustworthiness studies. Moreover, and as we will describe in detail in the following paragraph, the CFA results (see differences between the 3-factors and the 2-factor models) support the distinctiveness of the constructs.

Table 1 Descriptive statistics, reliabilities, and zero-order correlations
Full size table

Since ethical leadership, reputation for performance, and affective commitment to the organization were collected from the same source (i.e., subordinate), we tested the distinctiveness of the three constructs using confirmatory factor analysis (CFA). We compared our three-factor measurement model with two alternative, less-differentiated models (see Table 2): a two-factor model that combined the two evaluations of the supervisor (ethical leadership and reputation for performance); and a one-factor model where all the measures collected from the subordinate were combined. Lagrange multiplier tests (Bentler 1995) for adding parameters suggested adding two residual correlations to our model. The most differentiated model (three-factor model) presented the best fit (χ 2(204) = 400.09**; CFI = 0.90; RMSEA = 0.07; SRMR = 0.08). The Chi square difference tests demonstrated a better fit as the models became more differentiated (Δχ 2(2) = 22.14, p < 0.05 between the three-factor and two-factor model; and Δχ 2(1) = 196.53, p < 0.05 between the two-factor and one-factor model). Loadings ranged from 0.46 to 0.77 for ethical leadership, 0.34 to 0.79 for reputation for performance, and 0.36 to 0.86 for affective commitment to the organization. Thus, we used the hypothesized model in our analysis.

Table 2 Confirmatory factor analysis
Full size table

We tested our hypotheses using multiple regression and bootstrapping analysis since it is the most endorsed method to test conditional indirect effects, i.e., moderated mediation models (Preacher et al. 2007). Before creating the interaction term, we centered both predictors, i.e., ethical leadership and reputation for performance (Cohen et al. 2003). The results of our model are described in Table 3. Ethical leadership had a significant positive relationship with affective commitment to the organization (B = 0.58, p < 0.01), thus supporting hypothesis 1. Moreover, and while reputation for performance did not present a significant relationship with affective commitment to the organization (B = 0.12, p > 0.05), it significantly moderated the relationship between ethical leadership and affective commitment to the organization (B = 0.35, p < 0.01).

Table 3 Results of the moderated mediation analysis
Full size table

Following the recommendation of Cohen et al. (2003), we plotted the interaction effect. As predicted in hypothesis 2, the relationship between ethical leadership and affective commitment to the organization was stronger when reputation for performance was high (+1 SD; t = 6.78, p < 0.05) than when it was low (−1 SD; t = 3.74, p < 0.05). Although the relationship between ethical leadership and affective commitment to the organization remained significant when reputation for performance was low, the difference between slopes is significant (t = 4.26, p < 0.05).

Finally, and since affective commitment to the organization presented a significant negative relationship with organizational deviance (B = −0.15, p < 0.01), thus supporting hypothesis 3, we examined the indirect carryover effects of the interaction to organizational deviance through affective organizational commitment. When reputation for performance was high, the conditional indirect effect of ethical leadership on organizational deviance through affective commitment to the organization was significant (B = −0.11, 95 % CI [−0.2410 to −0.0439]). When reputation for performance was low, the conditional indirect effect was also significant, albeit smaller (B = −0.05, 95 % CI [−0.1505 to −0.0127]). These results support hypothesis 4. That is, as ethical leadership increases, subordinate’s organizational deviance decreases via (increased) affective commitment to the organization, particularly when subordinates rate their supervisors as having a high reputation for performance.Footnote 1

Discussion

This paper examined how (i.e., through an increase in affective commitment to the organization) and under what conditions (i.e., depending on reputation for performance) ethical leadership affects employee-deviant behaviors directed at the organization. Overall, the results supported our hypotheses. Ethical leadership was significantly related to affective commitment to the organization, particularly when supervisors also have a high reputation for performance. Furthermore, the conditional effect of ethical leadership on affective commitment to the organization (depending on the levels of reputation for performance) presented carryover effects for employee deviance directed at the organization. Taken together, these results contribute to the literature on ethical leadership by examining two understudied topics: how ethical leadership is associated with employees’ behaviors and the boundary conditions that affect that relationship.

Our findings suggest that ethical leadership contributes significantly to the development of the EOR through social exchange mechanisms. Given the attention other mechanisms have received in the literature (i.e., work environment, job characteristics), our research highlights the need to further examine the implications of ethical behaviors for social exchanges. The demonstration of normatively appropriate conduct through personal actions and interpersonal relationships (Brown et al. 2005) by supervisors provides cues concerning the values of the organization, thus enhancing employees’ affective organizational commitment. Thus, our results provide evidence not only that employees need ethical guidance from leaders, another of the main misconceptions Brown (2007) highlighted about ethical leadership, but also that they make inferences about the organization’s intentions based on the supervisor’s actions, one of the visible features of the organization’s anthropomorphization (Levinson 1965).

Furthermore, we found that the strength of the relationship between the supervisor’s ethical conduct and employees’ commitment to the organization varied according to the supervisor’s reputation for performance. Specifically, the impact of ethical leadership on the EOR is amplified for leaders that are also evaluated as high performers. Personal reputation is an important asset, because it reduces uncertainty concerning future behavior by signaling that the individual is trustworthy and has the necessary abilities to deal with workplace demands (Laird et al. 2009). This interaction effect is aligned with Hochwarter et al.’s (2007) claim that in order to fully understand how influence processes take place, one needs to examine the role of personal reputation.

Interestingly, if leaders are not perceived as ethical, regardless of their personal reputation for performance, employees show the lowest levels of affective commitment to the organization. It appears ethical leadership is a necessary but not sufficient condition for employees to demonstrate the strongest emotional bond with the organization. That is, when defining their relationship with the organization, employees examine both sides of the supervisor’s “means to an end” equation as an agent of that organization. It is as if employees ask two questions: Does this supervisor’s work conduct follow higher moral standards (i.e., does (s)he have high ethical leadership? If so, does (s)he also demonstrate high work effectiveness and achieve the proposed ends (i.e., does (s)he have a high reputation for performance)?

Lastly, we found that ethical leadership was indirectly related to employee deviance through an improvement in the EOR, measured by the increase in affective commitment to the organization. As Brown (2007) put it, ethical leaders encourage employees to aspire to do the right thing, which is more than simply obeying the law, and are quite effective at doing it. These results are also aligned with the norm of reciprocity (Gouldner 1960), in that employees feel obligated to reciprocate the higher standards (both ethical and for performance) put forth by the organization’s representatives, in this case the supervisor, through their own actions.

Our study also carries implications for practice. Our results demonstrate that promoting ethical leadership has an impact that goes beyond shaping ethical conduct (Brown et al. 2005). Ethics is a process (Brown 2007), and organizations and managers should be aware that ethical leadership influences how employees view the organization and determines their actions in the workplace. Organizations that aspire to promote a positive EOR and minimize deviant behaviors should, therefore, invest in the promotion of high ethical standards. A lot of attention has been given by the media to the (un)ethicality of top management teams and CEOs. Our study reinforces the idea that organizations also have to worry about the moral standards of those in direct supervision positions, as they are important role models due to their ability to reward or punish subordinates (Brown et al. 2005). Thus, organizations should make clear that they do not tolerate individuals—at any level of the hierarchy—that play the game close to the foul line, as Keough (2008) put it.

Organizations can effectively promote ethical leadership in multiple ways. They can, for example, measure applicants’ ethical standards at the time of entry in the organization and/or examine their actions in exercises or solving ethical dilemmas, selecting only those that are aligned with the set of moral standards held by the organization. They can also provide training to their line and middle-level supervisors, highlighting the organization’s core values and members’ expected actions, and explaining how leaders may convey those norms of conduct to their subordinates (e.g., inviting honest feedback or reinforcing employees that did “the right thing”, regardless of the outcomes: Brown 2007). Finally, they can also develop a corporate culture based on values with a strong emphasis on ethics. This culture should be consistently communicated across all levels of the organizations, in order to make sure: (a) each member is aware of the core ethical values of the organization; and (b) unethical behaviors are quickly and severely sanctioned.

Additionally, we found that ethical behaviors combined with a high reputation for performance produce the highest levels of affective commitment to the organization and the lowest levels of organizational deviance. While ethics are important, it appears individuals put that in perspective and search for additional signals concerning the leader’s intentions: following ethical guidelines but not being able to achieve goals provides a rather limited contribution to a positive EOR (and consequently to the employees’ reciprocation efforts). Thus, organizations should make an additional effort to retain those supervisors that present both features. This would facilitate their role modeling function, by signaling that both ethics and performance combined are key features to achieve status and success in the organization.

Limitations and Directions for Future Research

Despite the promising findings, our research is not without limitations. The first limitation concerns the cross-sectional nature of the study. All variables were collected in one single moment in time; thus, we should not make any causality inferences. Although our conceptual model was grounded in theory (and particularly in social exchange theory) and driven by past empirical research, our results should be interpreted with this in mind, and future research should attempt to test the ethical leadership process using longitudinal designs. A second limitation is related to common method variance (CMV) since three of the variables in our model were collected from the same source. However, our concerns about CMV are minimized by two reasons: (a) our model includes a variable (organizational deviance) evaluated by a different source (supervisor), which was significantly related to employee variables; (b) CMV does not augment interaction effects—it deflates them, making it more difficult to detect them (Siemsen et al. 2009).

Our study also opens up new avenues for research. Future research should continue to focus on the mechanisms linking ethical leadership and employees’ behavior, particularly how multiple mechanisms operate simultaneously. Research so far has examined these mechanisms (e.g., work environment, job characteristics, employee attitudes) separately, but in order to have a more complete understanding of how these processes take place, we need to re-examine them simultaneously. For example, as far as we know, there may be multiple paths to achieve the same goal (e.g., ethical leadership may be related to performance because it provides employees with a role model, because it puts in motion the norm of reciprocity or both) and there may be different paths to achieve different goals (e.g., ethical leadership may help employees deal with major organizational changes or affect well-being because it reduces uncertainty, but uncertainty reduction may not be a key mechanism for the development of an emotional bond with the organization).

In a similar vein, another potential line of research concerns the multi-foci conceptualization of social exchanges (Cropanzano and Mitchell 2005). A number of papers have underlined that, in order to have a complete perspective of how social exchanges operate, researchers should examine multiple social exchange foci simultaneously. For example, the studies conducted by Becker and his associates (Becker 1992; Becker et al. 1996) show that commitment to other foci, such as top management, supervisor, and work group, influences employee attitudes and behaviors over and beyond commitment to the organization. This multi-foci perspective of social exchanges has been recently applied to ethical leadership. The study developed by Hansen et al. (2013) indicates not only that ethical leadership at different levels (organizational and supervisory) helps develop commitment to the same foci through different processes, but also that within-foci effects are significantly stronger than cross-foci effects. Thus, future research should attempt to develop models of ethical leadership that integrate multiple foci of ethical leadership and its specific antecedents and consequences, as well as examine both within and across foci effects.

Researchers should also look into other potential boundary conditions (i.e., moderators), namely for cross-foci effects. Eisenberger et al. (2010) recently found that the generalization of supervisor’s actions to the organization as a whole depends on the extent to which employees identify their supervisor with the organization (supervisor’s organizational embodiment: SOE). In their two studies, they found that the strength of the relationship between leader–member exchange and affective commitment to the organization varied significantly depending on the levels of SOE. This raises an important question: Do employees generalize their perceptions of supervisor ethical behavior to the organization even when the supervisor is not aligned with the organization’s values and goals (i.e., has a low SOE)? Due to its implications for management, the role of SOE in the ethical leadership process deserves further examination.

In conclusion, our study is among the few to examine both the mechanisms and boundary conditions that link ethical leadership and employee behaviors. Overall, we found that ethical leadership affects the social exchange process between the employee and the organization (with consequences for employee deviance) and that this relationship is stronger when supervisors also have a high reputation for performance. While we believe our research significantly contributes to the discussion of the implications of (un)ethical behavior in organization contexts, much more research is needed to fully understand this phenomenon that, although recognized as a key feature of effective leadership, is still surrounded by alarming statistics.