Leadership has critical implications for groups, organizations, and societies. When it succeeds, its constituents prosper. When it goes wrong, teams lose, armies are defeated, organizations falter, and societies suffer. The bankruptcies of Enron and Worldcom, the tragic events at Jonestown, the scandals at Penn State and in the Catholic Church, and the widespread poverty in Germany after the fall of Hitler all highlight the destructive potential of leadership on organizations of various forms and purposes. Yet, when destructive leadership episodes occur, news headlines often focus on leaders, rather than the group processes and the larger historical, institutional, and societal forces that also contribute to the outcomes.

Even within the field of leadership studies, researchers have not been immune to leader-centric views. Indeed, the term “destructive leadership” has increasingly been used as an overarching expression for various “bad” leader behaviors (e.g., hostility, coercion, theft, corruption) believed to be associated with negative consequences for followers and/or the organization (e.g., Einarsen et al. 2007; Ferris et al. 2007; Krasikova et al. 2013; Shaw et al. 2011; Schyns and Schilling 2013). The term is found in articles comprising special issues of academic journals (e.g., Leadership Quarterly, vol. 18 2007), edited books (e.g., Schyns and Hansbrough 2010), and symposia at professional conferences (e.g., “Destructive Leadership: Measurement, Antecedents, and Outcomes,” SIOP 2009).

Yet, a more holistic understanding of destructive leadership requires recognition that leadership processes and their outcomes are rarely the product of a single factor or person (Lieberson and O’Connor 1972; Meindl et al. 1985; Salancik and Pfeffer 1977). Indeed, there is an appreciation in the broader leadership literature that the term “leadership” has been too narrowly defined, and that it reflects a dynamic, cocreated process between leaders, followers, and the environment (Avolio 2007; Hernandez et al. 2011). Over time, the confluence of these elements contributes to group, organizational, and even societal outcomes that vary in their constructiveness or destructiveness.

This paper focuses on destructive leadership processes and the harmful outcomes they create for organizations and their constituents. We make three key contributions to the literature. First, we offer a critique of leader-centric conceptualizations of destructive leadership. Expanding on and clarifying prior discussions (i.e., Padilla et al. 2007; Thoroughgood et al. 2012), we argue that, despite recognition of a broader leadership process in the larger leadership literature, current leader-centric definitions of destructive leadership continue to focus on “bad” leader behaviors. More specifically, although we as researchers claim to acknowledge the roles of followers, environments, and time in leadership processes, current definitions of destructive leadership largely overlook these factors.

Second, drawing on systems, institutional, and organizational ecology theories, we offer a new, broader definition of destructive leadership that is grounded in the current leadership discourse. While our intention is not to diminish the role of “bad” leader behaviors, we argue that theory and research will profit from a more balanced, integrative view, one that recognizes destructive leadership as a cocreated process between leaders, followers, and environments over time and that better captures organizational realities. Specifically, an inclusive definition should acknowledge destructive leadership as (a) a group process involving flawed, toxic, or ineffective leaders, susceptible followers, and conducive environments, and consisting of (b) destructive group or organizational outcomes, as well as (c) a dynamic time frame.

Third, based on a review of the literature, we identify areas of conceptual inconsistency and gaps in our understanding of leaders, followers, and contexts as they relate to destructive leadership processes. In terms of leaders, we discuss the role of intentions, negative influence versus counterproductivity, and active versus passive behaviors. With respect to followers, we underscore the need to consider follower susceptibilities, the cocreational roles of different followers in destructive leadership processes, and the developmental trajectories of these followers over time. Finally, we discuss environmental elements at multiple levels, including institutional (internal and external checks and balances), macroenvironmental (social, economic, and technological conditions), and cultural (societal attitudes and beliefs) influences, which are often ignored, but which critically shape destructive leadership processes and their outcomes.

Leader-Centrism and Destructive Leadership

The leadership literature has traditionally been leader-centric, focusing on traits and behaviors related to leader emergence (“Does this person look like a leader?”) and perceived effectiveness (“Is this person doing a good job?”) (Kaiser et al. 2008). A smaller body of research examines how leaders shape group processes (“How did the team play?”) and fewer assess group outcomes, or how a leader’s group performs (“Did the team win or lose?”). Bearing in mind the general belief that leadership reflects a group process that involves social influence to achieve group goals (Hogan et al. 1994), some have pointed out that the literature tells us more about how leaders are regarded than about whether their groups actually perform well and reach their goals (Kaiser et al. 2008). This focus on perceptions of leaders neglects the reality that leaders who are generally liked may be associated with poor performing teams and organizational decline (“bad” leadership outcomes), while leaders who are generally disliked may be associated with high-performing teams and organizational success (“good” leadership outcomes).

Moreover, even when group processes and group outcomes are considered, the critical roles of followers, the environments in which leaders and followers interact, and time are frequently overlooked. Despite recent advancements (e.g., Carsten et al. 2010; Uhl-Bien et al. 2014), followers are often viewed as passive recipients of leaders’ influence and instruments for attainment of a leader’s goals, while the environment tends to be treated as a moderator of the effects of leaders on followers. Moreover, most studies ignore the role of time, masking changes in trajectory that leadership processes may take over time (Shamir 2011). Thus, even though we claim to appreciate these factors, we often fail to truly integrate them into our definitions and analyses of leadership phenomena.

Similarly, existing perspectives on destructive leadership are primarily leader-centric, focusing on traits and behaviors believed to produce “destructive” consequences for followers and organizations. Traits comprise, among others, narcissism, Machiavellianism, and a personalized need for power (e.g., House and Howell 1992; Rosenthal and Pittinsky 2006). Behaviors have fallen under various overlapping follower-directed concepts (see Thoroughgood et al. 2012), including petty tyranny (Ashforth 1994), abusive supervision (Tepper 2000), and supervisor undermining (Duffy et al. 2002) and, to a lesser degree, organization-relevant concepts, such as toxic leadership (Lipman-Blumen 2005) and leader derailment (McCall and Lombardo 1983). While the former tend to include perceptions of coercion, abuse, and arbitrariness, the latter tend to include reports of, among other things, corruption, sabotage, and theft (Einarsen et al. 2007; Thoroughgood et al. 2012b) (see Table 1). Although leader traits and behaviors are relevant and worthy of study, we argue that they alone do not capture the whole “story” of destructive leadership processes nor do they ensure that destructive leadership outcomes will occur.

Table 1 Concepts associated with research on destructive leadership
Full size table

Explaining Leader-Centrism

Why do so many articles and stories stress the role of leaders in destructive leadership episodes and often overlook the roles of followers and environments in contributing to the results? First, there is a fascination with leadership outcomes, particularly destructive ones. Brutal dictators, political scoundrels, and deceitful CEOs invite speculation on the “dark” traits related to organizational toxicity and decay. It is unsettling to consider that those who lead firms, political institutions, universities, and religious groups may also be narcissistic psychopaths, despite being able to create some positive change. When disastrous outcomes occur, we often fail to ask, “What factors, in addition to the leader, contributed to the results?”

A second reason is a popular perception of leadership that looks to leaders for answers to group and organizational problems. For instance, romantic views of leaders tend to attribute unequal weight to their impact on group outcomes (Meindl et al. 1985). Meindl’s research confirmed a human tendency to extoll leaders when an organization succeeds and blame them when it fails; yet, Meindl warned against leader-centric explanations. Relatedly, members of individualistic societies tend to make more leader-based attributions, even when followers and situations are clearly involved in the results (Oyserman et al. 2002). Individualists are socialized into defining people as individual units; as such, they are likely to attend to person-centered characteristics when explaining behavior and outcomes (Morris and Peng 1994).

Third, much of the leadership literature is a reflection of psychologists’ traditional emphasis on traits and behaviors, as opposed to higher macro-level processes more inherent to fields like sociology, institutional economics, history, and political science. Since the 1930s, psychologists have developed many of the guiding theories and methods shaping leadership research (House and Aditya 1997). While the contribution of trait and behavioral analyses should not be minimized, one might ask whether our understanding of leadership might be more complete today had the field started with a broader focus. At this stage, we argue that researchers who study destructive leadership should ask a similar question.

Finally, the simultaneous analysis of leaders, followers, and environments is difficult given all the factors to consider. It is much easier to utilize traditional surveys that measure perceptions of leaders. As Hunt and Dodge noted, “Questionnaires seem to be with us always. They are just too quick and easy” (2001, p. 454). Still, Hunt and coworkers (Hunt and Dodge 2001; Hunt and Ropo 1997) argued that such challenges do not excuse quick, one-shot studies nor should they preclude the use of broader systems approaches.

Two General Problems with Leader-Centric Definitions of Destructive Leadership

There are two general problems with leader-centric definitions of destructive leadership. First, these definitions assume that certain leader behaviors are sufficient for destructive leadership outcomes to occur, regardless of whether they eventually result in any serious damage to the group or not. This overlooks the possibility that adequate checks and balances (e.g., internal oversight committees, external regulatory bodies) may remove a leader before he or she can seriously harm the group or organization; that followers may oppose dysfunctional leaders and prevent long-term damage to the organization and its constituents; or that some “bad” leader behaviors (e.g., aggression, autocratic decision-making) might even lead to gains for some organizations and their members in certain contexts. For example, based on current definitions, leadership under former NCAA basketball coach, Bobby Knight, would largely be considered “destructive” due to his aggressive style of leading. This is despite the fact that Knight led the Indiana Hoosiers to 3 National and 11 Big Ten championships, won 661 games (0.733 % winning percentage) at Indiana, had a player graduation rate of 98.0 %, and is revered by most of his players for the life skills he taught them (Feinstein 2012). Similarly, despite his perfectionism and callousness, Apple’s former CEO, Steve Jobs, was a strong force behind Apple’s immense success during his tenure from 1997 to 2011. In sum, it is difficult to link “bad” leader behaviors clearly with destructive leadership outcomes across all contexts.

Second, and relatedly, leader-centric conceptualizations do not incorporate the roles of followers and environmental conditions in destructive leadership processes. They do not address why certain types of followers are vulnerable to flawed, or toxic, leaders, how they shape their leaders’ motivations, self-images, and behaviors, or why other types of followers consciously participate in destructive leadership episodes. They also do not address how conducive environments influence and are influenced by “bad” leaders and dysfunctional leader–follower relationships. Thus, leader-centric definitions do not address the roles of followers and environments in explaining why destructive leadership processes happen or why they persist long enough to produce destructive outcomes for organizations and their constituents.

To be clear, our intention is not to excuse or to make any moral judgments about certain “bad” leader behaviors. Rather, we seek to present a pragmatic view on the limitations of focusing solely on these behaviors in order to spur a broader discussion of destructive leadership in organizations. Thus, the choice of acceptance or rejection is left with the readers. In the following sections, we discuss more specific challenges related to the two general problems discussed above and delineate a more inclusive definition of destructive leadership that addresses these problems. However, we first introduce systems, institutional, and organizational ecology theories to provide a theoretical foundation for our discussion.

Systems, Institutional, and Organizational Ecology Theories

To understand better the importance of previously overlooked elements of destructive leadership processes, we draw on systems, institutional, and organizational ecology theories. These theories provide a useful lens to begin developing a more complete perspective on destructive leadership in organizations.

Systems theories (Katz and Kahn 1978; Von Bertalanffy 1968, 1972; Senge 1990; Weick 1979) maintain that outcomes at the collective level are a product of the interactions of individuals comprising the system. These theories emphasize the interdependence of personnel in organizations, as well as the impact of external environments on organizational structures and functions (Kast and Rosenzweig 1972). Individuals reflect embedded components of a broader interdependent collectivity; structured roles are assumed to be interrelated such that they form a network of interconnections and reciprocal relationships among people occupying them. What this means for leadership processes is that leaders and followers are constrained and influenced by one another and by internal (e.g., organizational rules, policies, etc.) and external (e.g., government laws, regulations, etc.) environmental conditions. Systems theories thus remind us that leadership outcomes, whether constructive or destructive, are seldom a sole reflection of individual leaders and their behaviors. Rather, it is the interlocking nature of systems and interactions between leaders and followers at multiple levels within a particular environment that must be examined.

Institutional theories provide a complementary approach to understanding organizations and the behavior of their members, including leaders and followers. They consider the processes through which organizational structures, such as norms, rules, routines, and schemas, become established guidelines for social behavior (Scott 1987). These theories have several variations. Particularly relevant to our analysis, one set of theories maintains that, through ongoing interaction, a social order develops in organizations based on members’ acceptance of a shared social reality (Berger and Luckmann 1967). This conception is taken for granted as defining the “way things are and/or should be done” and, in turn, leads to repeated patterns of behavior (Zucker 1977, 1987). Specifically, social order is a product of individuals acting, interpreting their actions, and sharing their interpretations. These interpretations, or “typifications,” reflect attempts to classify behaviors and actors into categories (e.g., leaders dictate orders, subordinates follow them), which individuals use to respond to behaviors in a similar fashion (Berger and Luckmann 1967). Institutionalization reflects the process through which behaviors become repeated over time and ascribed similar meaning by individuals. Similar to a systems view, institutional theories would suggest patterns of dysfunctional, or toxic, leader behavior and their outcomes cannot be analyzed in a vacuum by focusing only on leaders. Rather, they would suggest such behaviors must be analyzed via a process-based lens, one that considers the social environment (e.g., followers, organizational norms and history) in which they, over time, become accepted and ingrained in the underlying social fabric of organizations.

Finally, from a more macro-level view, organizational ecology focuses on explaining how social, economic, historical, and political conditions impact the birth, survival, and decline of organizations and their structure over time (Hannan 1993). It views long-term changes in organizations’ internal “forms” as a result of selection processes (Aldrich 1979), whereby structural inertia undermines organizational adaptation in the face of shifting environmental demands. Organizations unable to adapt are replaced, or “selected out,” by those better suited to meet such demands. Specifically, changes at the population level are a product, in part, of variations, or any intended or unintended changes that organizational members create via their efforts to adjust to other members and to alter the organization’s relationship to the environment (Baum and Amburgey 2000). For example, the strategic initiatives of leaders represent variations. In turn, certain variations, whether planned or accidental, prove more beneficial in addressing changing conditions and are thus “selected for” by the environment, while others are not and are thus “selected out” (Hannan and Freeman 1977). As such, organizational survival and failure depend on the interaction between an organization’s internal “form” and environmental conditions (Freeman 1982).

Applied to leadership, an ecological perspective suggests that the environment is ultimately what determines the outcomes of leadership processes. As such, it stands in contrast to traditional adaptation views on organizational survival (Astley and Van de Ven 1983), which focused on the strategic decisions of leaders in shaping their organizations’ relationship to the environment. Such views stressed the traits, skills, and abilities of leaders in shaping organizational success or decay, deflecting attention away from environmental forces, internal and external to organizations, which also influence leadership outcomes (Baum 1999). Put simply, ecological theories remind us that while leaders do impact their organizations’ futures, environments play a powerful role in constraining leader–follower activities and thus leadership outcomes. Just as environments can deter “bad” leaders and followers from creating destructive results (e.g., when government agencies adopt greater regulations on organizational activities), environments can also overwhelm “good” leaders and followers to the point at which “destruction” ensues (e.g., when industries are uncertain, limiting leaders’ ability to develop and enact effective organizational changes).

Taken together, although systems, institutional, and ecological theories differ in their focus and scope with respect to explaining the complexities of organizational life, each points to the importance of adopting a broader, contextualized understanding of leadership processes and their outcomes—one that includes the vital role of leaders, but also the environments in which they operate over time. In terms of destructive leadership processes, systems and institutional theories would emphasize that dysfunctional leader behaviors and their outcomes cannot be analyzed in isolation from the environments, internal and external to organizations, in which such behaviors are shaped and reinforced over time. From a macro-level view, ecological theories further suggest that even well-intentioned leaders can be associated with organizational “destruction” due to the constraints that uncertain environments place on leaders to devise and implement changes fast enough to meet changing demands. Unfortunately, few attempts have been made to use these theories to develop a broader perspective on leadership phenomena (Wielkiewicz and Stelzner 2005). Below, we use these theories to derive a more balanced view of destructive leadership.

Toward a More Holistic Definition of Destructive Leadership

Drawing on the underlying principles of systems, institutional, and ecological theories, as well as more integrative approaches in the leadership literature, we suggest that more holistic conceptualizations of destructive leadership should explicitly include followers, environments, and time. In our view, destructive leadership reflects a special case of more general leadership situations, with the key difference residing in the degree to which the behaviors of flawed, toxic, or ineffective leaders (i.e., individuals with certain traits and characteristics) interact, over time, with followers and environments that are weak, susceptible, or conducive, leading to aggregate destructive outcomes for groups, organizations, and their constituents. Specifically, we define destructive leadership as a complex process of influence between flawed, toxic, or ineffective leaders, susceptible followers, and conducive environments, which unfolds over time and, on balance, culminates in destructive group or organizational outcomes that compromise the quality of life for internal and external constituents and detract from their group-focused goals or purposes. Our definition lends itself to most, if not all, contexts—from businesses and nonprofits, such as universities and government agencies, to religious groups to political and military institutions. It incorporates three essential features of destructive leadership: group processes, group outcomes, and a dynamic timeframe (see Fig. 1).

Fig. 1
figure 1

A more holistic conceptualization of destructive leadership processes

Full size image

First, consistent with systems, institutional, and ecological theories, as well as more integrative views in the leadership literature, this definition goes beyond the specific traits and behaviors of leaders. We do not define destructive leadership as a “bad” leader or as something that is done, consciously or unconsciously, to followers (e.g., hostility) or the organization (e.g., theft). Rather, we view destructive leadership as a social, or group, process that involves interactions between flawed, toxic, or ineffective leaders, susceptible followers, and conducive environments. Leader actions or inactions are a part of these processes, but not these processes alone. As such, we depart from definitions that view destructive leadership solely in terms of leader behaviors (e.g., Einarsen et al. 2007; Krasikova et al. 2013; Schyns and Schilling 2013).

Second, because leadership is a group process (Avolio et al. 2003; Hollander 1964), it involves group outcomes (Kaiser et al. 2008). We believe the concept of destructive leadership should be based on a similar understanding. In our view, the extent to which leadership processes, in their entirety, are “destructive” should be determined based on the degree to which they, by and large, harm the welfare of the group they are meant to serve, not whether certain leader behaviors are viewed negatively by some followers. As such, destructive leadership entails negative outcomes to the group, with certain processes between leaders, followers, and environments being more likely to result in these outcomes than others.

Third, our definition implies a dynamic time frame. Destructive leadership is typically not a static phenomenon that can be captured via cross-sectional accounts of leader behavior. Leadership processes change trajectories over time depending on the evolving interactions among leaders, followers, and the environment. Thus, they are seldom entirely “constructive” or “destructive”; they involve outcomes that fall along a constructive-destructive continuum. Determining whether a leadership process is largely “destructive” requires an examination of whether it resulted in outcomes that, on balance, were harmful to a group once it has exhausted its course (e.g., a CEO stepping down, a president’s term ending, a coach retiring). As our earlier examples of Bobby Knight and Steve Jobs highlight, whether one agrees with their styles or not, it is tough to argue, upon reflection, that the leadership episodes they presided over were destructive in their totality. We discuss these features of our definition in greater detail below.

Feature 1: A Social or Group Process

As Avolio (2007) noted, “[Understanding leadership phenomena] requires an examination that considers the relevant actors, context (immediate, direct, indirect), time, history, and how all of these interact with each other to create what is eventually labeled leadership” (p. 25). Similarly, other writers have pointed out that the terms “leader” and “leadership” are not the same and that leadership processes involve more than the behaviors of leaders (see Table 2). These more contemporary views on leadership align with systems and institutional theories, which again focus on the interactions between a system’s interrelated parts rather than on one part in isolation. As such, these theoretical traditions both maintain that organizational phenomena represent dynamic social processes (Astley and Van de Ven 1983; Scott 1987). In our view, the concept of “destructive leadership” should be based on a similar understanding. Indeed, despite the pervasiveness of leader-centric views, broader perspectives on leadership phenomena are not new. For example, the contingency theories (e.g., House 1971; Fiedler 1964) sought to identify how characteristics of followers and the environment impact a leader’s influence on followers. Although these theories made important strides in incorporating followers and environments into the leadership discourse, they focused largely on leaders’ effects on followers, rather than on the interactions between leaders, followers, and environments that comprise leadership processes and that shape their outcomes.

Table 2 Notable criticisms of leader-centric approaches
Full size table

Recently, there has been a growing trend toward more integrative approaches that do not define leadership phenomena in terms of leader behaviors. For example, social identity theory (Hogg 2001) defines leadership as a group process, whereby a leader’s endorsement and influence depend on whether he or she is seen as prototypical of the group’s identity. Echoing a systems and institutional perspective, complexity theory (Uhl-Bien et al. 2007) distinguishes between leaders and leadership, suggesting the latter is an interactive process created by networks of interdependent agents embedded in context (e.g., political, historical, organizational). As Uhl-Bien et al. (2007) noted, “Leadership is too complex to be described as only the act of an individual or individuals…it is a complex interplay of many interacting forces” (p. 314). Shared and distributed theories (e.g., Gronn 2002; Pearce and Conger 2002) also define leadership as a process, not a behavior or set of behaviors of a particular leader. Importantly, these more integrative views underscore the multilevel nature of leadership phenomena. As Yammarino and Dansereau (2011) noted, failing to integrate micro- and macro-levels of analysis leads to an incomplete understanding of leadership phenomena, leading to faulty measures, improper analytic techniques, and invalid conclusions. They argued that theory building and testing can only be advanced if higher levels are explicitly viewed as the context for, or boundaries on, lower levels (i.e., individual behaviors occurring in group settings; individual and group behaviors occurring within a broader organizational context) (see also Johns 2006).

Understanding destructive leadership requires a similar, more integrative perspective. Defining destructive leadership in terms of “bad” leader behaviors is useful for clarifying the leader component of destructive leadership processes. Yet, as systems, institutional, and ecological theories would together suggest, it does not capture other important elements of these processes, including followers’ reactions to “bad” leader behavior, the evolution of dysfunctional leader–follower relationships over time, or the embeddedness of these relationships in the broader historical, institutional, industry, and societal context. As an example, Krasikova et al. (2013) recently conceptualized destructive leadership as:

Volitional behavior by a leader that can harm or intends to harm a leader’s organization and/or followers by (a) encouraging followers to pursue goals that contravene the legitimate interests of the organization and/or (b) employing a leadership style that involves the use of harmful methods of influence with followers, regardless of justifications for such behavior (p. 1310).

According to this definition, such behaviors of leaders are, by their nature, sufficient to produce destructive leadership. However, what if individuals (a) refuse to follow, or actively oppose, the leader’s orders and/or (b) are unaffected by the leader’s methods of influence? By focusing solely on the leader’s behavior and overlooking followers and their reactions to such behavior over time, this definition makes it hard to determine (a) who is being “led” and (b) what is being “destroyed.” Further, what if (a) checks and balances, internal or external to the organization, remove the leader from power before he or she can seriously damage the long-term performance of the organization and the wellbeing of its members, or if (b) the organizational, industry, or societal context is one in which certain “bad” leader behaviors (e.g., aggression, unilateral decision-making) are expected or even necessary for organizational performance and survival (e.g., industries marked by uncertainty, military settings, high power distance societies)? By overlooking the environmental context, this definition assumes that certain leader behaviors will inevitably result in “destructive” effects for all organizations and their constituents, regardless of whether they do so or not. Below, we elaborate on these points. Yet, consistent with systems, institutional, and ecological theories, we highlight some of these simplistic assumptions of leader-centric definitions to point out broader follower and environmental factors that must be considered when defining destructive leadership.

Bearing in mind limitations of leader-centric approaches and in concert with the growing trend toward more integrative views on leadership phenomena, we maintain that destructive leadership is not simply a “bad” behavior or set of behaviors of a leader, but rather involves complex interactions among flawed, toxic, or ineffective leaders and susceptible followers, which unfold over time within multiple layers of context. Thus, in our view, followers’ conformity to or active collusion with “bad” leaders are necessarily a part of destructive leadership processes, as are broader institutional factors (e.g., a lack of checks and balances), macro-environmental conditions (e.g., crises, industry uncertainty), and societal forces (e.g., power distance) that shape, reinforce, and allow dysfunctional leader–follower relationships to occur over time and, in turn, cause serious harm to the long-term performance of the group and the welfare of its constituents. Again, this broader perspective does not minimize “bad” leader behaviors. Rather, it alludes to a broader set of processes that occur between “bad” leaders, susceptible followers, and conducive environmental conditions that, together, may eventually be labeled “destructive leadership”.

By adopting a broader lens, we can begin to understand the complex dynamics underlying real-world cases of destructive leadership. For example, whether we examine destructive leadership episodes in religious groups (e.g., The People’s Temple, Branch Davidians), political dictatorships (e.g., Stalin’s Russia, Hitler's Germany, Pol Pot’s Cambodia), or corporations (e.g., Enron, Worldcom), what is clear is that such episodes all involved followers who turned a blind eye, obeyed, or actively colluded with “bad” leaders, as well as environments that allowed such leaders to assume power and pursue their goals. As a specific example, under Idi Amin, the “Butcher of Uganda,” 300,000 people were slaughtered in a campaign of genocide and elimination of Amin’s rivals. Under his orders, military officers conducted public executions, while secret police tortured and murdered thousands of suspected dissidents. Further, Amin’s reign cannot be viewed in isolation from the factors that brought his regime to power and that allowed it to remain for so long. For instance, while Uganda was still a British colony and despite Amin’s penchant for cruelty as a soldier, he was promoted by the British to “Afande,” the most powerful position a Black African could hold in the colonial army. This provided Amin the chance to seize power in a military coup in 1971. The regime’s ability to secure power was enhanced by the public’s attraction to Amin’s charisma and naïve belief that the military government would remain only until elections could be held. Eight years later, Uganda had been transformed into a state riddled by ethnic persecution, human rights abuses, and gross economic mismanagement. Like other cases of destructive leadership, this example is illustrative given it highlights the process by which a flawed leader, working in unison with susceptible followers and an environment marked by defective institutions, can have devastating outcomes for even whole societies.

Taken together, although leader-centric definitions of destructive leadership are useful, they do not account for the expansiveness of these processes within organizations. As systems, institutional, and ecological theories would suggest, followers and environments must also be considered. By shifting the focus toward a more holistic view, a more realistic understanding of destructive leadership is possible.

Feature 2: Destructive Group or Organizational Outcomes

Defining destructive leadership also requires consideration of outcomes. Like other leadership concepts, destructive leadership is typically defined and assessed in terms of the average reported effects of certain “bad” leader behaviors on subordinate criteria (e.g., job satisfaction, citizenship behavior). Yet, few studies consider group or organizational outcomes. Indeed, Kaiser et al. (2008) found in a review of ten meta-analyses that only 18 % of studies in the leadership literature utilized group or organizational outcomes as criteria. In pointing out this weakness, Kaiser et al. (2008), like others (e.g., Lord et al. 1984; Hogan et al. 1994), underscored the critical distinction between how leaders and their behaviors are perceived (i.e., how individuals generally feel about or the extent to which they approve of a leader) and how their teams and organizations actually perform. They noted that while understanding how leaders are perceived provides useful information about individual leaders, it does not tell us about the actual outcomes of leadership processes (i.e., whether the group ultimately achieved its goals and was better off because of a leadership process or not), which again are a product of more than leaders. From this perspective, it is not how individual leaders are regarded by certain subordinates that matters when evaluating leadership, but rather how their groups and organizations actually perform over time.

Echoing this view, Hogan and Kaiser (2005) stated that “Leadership is about the performance of teams, groups, and organizations; good leadership promotes effective team and group performance, which enhances the wellbeing of incumbents; bad leadership degrades the quality of life for everyone associated with it. [Because] leadership is a collective phenomenon, it follows that leadership should be evaluated in terms of the performance of the group over time” (pp. 169–172). Thus, leadership processes reflect functional tools for group performance; they involve influence in pursuit of collective enterprises that have consequences for a group’s long-term performance and, in turn, the welfare of its constituents.

Extending this logic to destructive leadership, it should be noted first that we do not diminish the importance of examining certain individuals’ subjective evaluations of “bad” leader behavior, including perceptions of hostility or intimidation, and their reported effects on individuals’ job attitudes, wellbeing, and behavior. Yet, consistent with Kaiser et al. (2008) and as our earlier examples of Bobby Knight and Steve Jobs further highlight, survey evaluations of “bad” leader behaviors tell us more about the leader being evaluated than they do about whether a leadership process harmed, by and large, a group’s ability to achieve its goals and left it worse off than before. Although Jobs rubbed some subordinates the wrong way, Apple flourished during his tenure, in part due to Jobs’ vision, but also many exceptional engineers and a company culture that demanded excellence. This is not to dismiss reports of mistreatment from some who worked under him. Yet, given leadership results are also a product of followers and environmental conditions, as well as functional acts of leaders that may co-occur with “bad” behaviors over time, just because some subordinates report certain “bad” leader behaviors does not ensure “bad” leadership outcomes will ensue.

Thus, from our perspective, destructive leadership has more to do with whether certain leaders, in conjunction with certain followers and environments, actually harm a group’s long-term performance and, in turn, the collective wellbeing of its constituents. From this view, Knight and Jobs presided over leadership episodes that were largely constructive for their groups, despite their personal difficulties in interacting with certain subordinates and stakeholders. In contrast, Enron, for example, was an instance of destructive leadership given corrupt leaders, with the aid of conforming and colluding followers, a lack of internal and external checks and balances, and a culture of greed, led the company into bankruptcy—degrading the quality of life for employees and investors alike. Concepts involving perceptions of “bad” leader behavior, such as “abusive supervision” or “petty tyranny,” may very well be a part of destructive leadership processes and deserve research. Yet, they do not capture the expansiveness of these processes.

Before proceeding to our definition’s final feature, several additional observations about leader-centric definitions should be noted in order to further highlight the utility of focusing on group outcomes as opposed to certain interpersonally “bad” leader behaviors. First, by defining and assessing destructive leadership in terms of these behaviors, it is assumed that they are, by nature, universally harmful to all followers (and presumably groups and organizations). However, leader–follower dynamics are seldom as clear-cut as implied by many analyses. Indeed, perceptions of and reactions to many of these behaviors may vary across different individuals. For example, expressions of anger are a part of measures of both “tyrannical” (Ashforth 1994) and “abusive” (Tepper 2000) leader behavior. However, studies suggest subordinates low on agreeableness may actually react positively to a leader’s displays of anger (Van Kleef et al. 2010). Such individuals expect less civility and are less sensitive to inconsiderate behavior (Graziano et al. 1996). As such, they may accept a leader’s hostility given social conflict is less distracting (Suls et al. 1998) and more motivating to them (Van Kleef et al. 2010). Further, some people underreport supervisor hostility, while others, such as those with negative affect or a hostile attribution bias, may overstate their exposure (Martinko et al. 2011; Tepper et al. 2006). Indeed, as Chan and McAllister recently pointed out:

Empirical findings in this domain have often been discussed and interpreted as evidence concerning abusive treatment of a more objective and independently verifiable sort. However, employee responses … are shaped by the mind-states of followers as well as the behavior of supervisors. Indeed, deliberative and attribution processes internal to employees have important roles to play in determining not only whether and to what extent … supervisory action and inaction are interpreted and understood as being abusive, but also the nature of employee affective, cognitive, and behavioral responses … (2014, p. 44)

Thus, one person’s definition of “abusive” behavior may differ from another person’s, making it difficult to objectively conclude that such behaviors will equally affect all people to the same degree or in the same ways. Further, one person’s toxic leader may be another person’s hero, depending on the needs of followers and their unique relationships with the leader (Lipman-Blumen 2005). For instance, some people are more likely to idolize charismatic leaders despite mistreatment given such leaders are able to satisfy their needs for security, meaning, and group membership (Curtis and Curtis 1993; Deutsch 1980). Further, subordinates with high leader–member exchange relationships with leaders have reported certain “toxic” behaviors (e.g., yelling when deadlines are missed, ridiculing an employee’s work) as less “toxic” (i.e., less likely to demoralize, upset, or leave an enduring effect on them) (Pelletier 2012).

Second, assuming that these behaviors always lead to harmful effects across all contexts appears to be problematic as well. Indeed, how such behaviors are defined, perceived, and reacted to depends on the social, cultural, and organizational context. For example, actions that are more generally regarded as “abusive” in low power distance societies may be less impactful and viewed as less unfair in high power distance societies where autocratic influence is more culturally normative (Hofstede 1984; Lian et al. 2012). Prior GLOBE results suggest autocratic leaders may even be preferable to democratic leaders in societies high on power distance, collectivism, masculinity, and uncertainty avoidance, including many nations in Eastern and Southern Europe, Asia, Central and South America, and the Middle East (Javidan et al. 2006). As such, what many Westerners perceive as “bad” leader behavior might be preferred or even necessary for group performance in other cultures. As Muczyk et al. noted:

Unqualified support for democratic [leaders] and individual autonomy has been the cultural norm of U.S. society since the end of World War I. Most of the post WWII leadership literature has been generated by American scholars. Consequently, the democratic predispositions found in the U.S. culture were assimilated into the leadership literature, consciously or otherwise. The predilection toward democracy is strengthened by the tendency for certain words, such as “autocratic” and “directive,” to take on a pejorative connotation irrespective of their denotation (Muczyk and Adler 2002, p. 3). Yet, when the entire corpus of leadership research is examined objectively, the unqualified acceptance of democratic [leaders] is largely a leap of faith as opposed to a conclusion based on empirical research, or for that matter, on logic and experience (Muczyk and Steel 1998, p. 40).

In discussing the paradoxical nature of “managerial tyranny,” Ma et al. (2004) similarly pointed out the role of culture in shaping responses to autocratic forms of leading, stating that:

Any analysis or discussion of “tyrannical” management styles…gets mixed up with idealistic considerations. For example, to most Americans, the governance style of many East Asian countries…is clearly inimical and antithetical to American ideals. But the fact remains that it is precisely those governance styles …that have transformed those countries into economic juggernauts. Seemingly, the vast majority of the populations in those countries are comfortable with the trade-offs between economic properity and personal freedoms. The experience of these countries illustrates…that there are certain contexts in which tyrannical styles of management produce needed results…Thus, the tyrannical behavior of the most celebrated leaders is paradoxical in nature and motivates us to develop a deeper understanding.

In addition to culture, perceptions of and reactions to these leader behaviors may also depend on organizational norms and conditions. For example, leader “abuse” might be defined very differently in the military, hospital emergency rooms, blue-collar industries (e.g., construction, trucking, logging, etc.), the NFL, and other high-stakes situations requiring swift decision-making and safety. In these and other contexts, aggressive leader behaviors can also build toughness, establish a chain of command, promote adherence to safety standards, and cultivate a collective identity among group members. This may mean the difference between group success or failure in the face of environmental turbulence and stress. More generally, during times of crisis, people often look to autocratic leaders to make difficult decisions and restore order (Janis and Mann 1977). Indeed, Beer and Nohria’s (2000) discussion of Theory E change initiatives, which stress shareholder value via drastic layoffs and restructuring, points to the harsh reality that autocratic leaders are sometimes needed for corporate turnaround efforts that require bold, time-sensitive decisions. It should be noted that this uncomfortable tension between two things we value and always want together, effectiveness and niceness, has been a point of discussion for centuries. Indeed, as Machiavelli argued long ago, leaders often must sacrifice kindness for practical effectiveness or ignore the feelings of some constituents to promote the greater good. Although Machiavelli believed in acting with virtue and being kind when possible, he also believed that a leader’s use of force and manipulative influence were, at times, inevitable costs of dealing with the world as it is, not how we feel it should be.

In sum, without minimizing certain individuals’ reports of and negative responses to specific “bad” leader behaviors, it is important to recognize that (a) not all subordinates within the same context or across different contexts will necessarily report or respond in the same ways to these behaviors; and that (b) “bad” leader behaviors, whether objectively real or subjectively perceived, do not ensure “bad” leadership outcomes for the group. Thus, in our view, destructive leadership is a matter of group results, not whether certain leader behaviors are viewed unfavorably by some followers. This is not to suggest leader behaviors are not a part of destructive leadership; they clearly are. But also, given leadership is a collective process that involves followers and the context, group results should be carefully considered.

Feature 3: Dynamic Time Frame

Despite the reality that leadership processes are dynamic and unfold over time, most leadership studies assume that observed relationships are not time-contingent (Hunter et al. 2007). Shamir (2011) noted that the dominant paradigm in leadership research, which suggests certain behaviors of leaders impact more proximal subordinate criteria (e.g., job motivation), is atemporal and overlooks that these leader inputs and their effects may change over time. For example, a coach’s “tough love” may initially be frustrating and demotivating to players. Yet, over time, they may come to appreciate the coach’s style and become motivated as a result of it. However, by overlooking the role of time, these dynamic effects on players and the team’s long-term performance are masked. This issue of time is exacerbated by the fact that leadership processes, consistent with systems and institutional theories, do not involve one-way influence but rather interactions between leaders and followers over time. These time-varying interactions, in turn, shape the long-term outcomes of leadership processes for groups and organizations. As such, given leadership processes change trajectories depending on the evolving interactions between leaders, followers, and the environment, their aggregate outcomes may not be apparent for some time.

To understand destructive leadership, a dynamic time frame must also be applied. As suggested above, determining whether a leadership process is “destructive” requires an evaluation of whether the process, in its totality, harmed a group or organization’s long-term performance and, in turn, the quality of life of constituents. Without taking a dynamic view, it is difficult to evaluate the “destructiveness” or “constructiveness” of most leadership episodes given their cumulative results take time to develop and tend to fall along a continuum, from absolutely terrible to absolutely great, with most falling somewhere in the middle. In fact, research and practical observation indicate that leaders produce both “good” and “bad” effects in the short term (Aasland et al. 2010; Rayner and Cooper 2003). For example, under the widely admired Roberto Goizueta, Coca-Cola became a top U.S. firm. Yet, Goizueta’s tenure as CEO is also associated with the widely publicized “New Coke” fiasco that cost the company millions. On the flip side, destructive leadership processes often involve some positive gains, at least in the short term. For example, even the most tyrannical regimes have brought some value to their constituents. Under Mussolini, Italy suffered widespread devastation to its military, economy, and architectural treasures due to his regime’s alignment with Hitler during World War II. Yet, in the regime’s early days, Italians profited from improved public transportation, public works development, and job opportunities, which provided stability, a sense of national pride, and relief from the economic and political crises of the time.

These observations point to the reality, as unsettling as it may be, that even highly constructive leadership episodes often involve misfortune for some followers and costly short-term setbacks for the group, while largely destructive leadership episodes may produce benefits for some followers and short-term gains for the group. Thus, in our view, defining a leadership process as “destructive” requires an evaluation of whether the process culminated in outcomes that were, on balance, harmful to the group or organization and, in turn, the collective goals and wellbeing of its constituents. If destructive leadership is defined in terms of outcomes that are, in their totality, destructive to the group, this leaves room for “good” leaders to create “bad” outcomes and “bad” leaders to create “good” outcomes in the short term.

Temporariness of Destructive Leadership Outcomes

Several observations are worth noting regarding leadership outcomes, especially with respect to destructive leadership. First, our definition does not suggest that a group, organization, or even society be entirely “destroyed” for a leadership episode to be labeled “destructive.” Most groups and organizations experience multiple leadership episodes across their lifespans that tend to vary in their constructiveness or destructiveness. For example, Nazi Germany under Hitler was clearly a case of destructive leadership, but Germany today is a thriving and powerful country once again. The point is that for many reasons, including mortality (e.g., leaders and followers dying), attrition (e.g., leaders and their groups retiring), term limits (e.g., leaders and their administrations coming to a preset end), or removal (e.g., leaders and their regimes being ousted from power), leadership outcomes, destructive or constructive, are temporary. Most groups, organizations, and civilizations have risen and fallen, prospered and declined, won and lost, and many have vanished or gone extinct—suggesting leadership episodes have a start, a life, and an end.

Second, although leadership episodes have a genesis, a life, and an eventual conclusion, of which the outcomes are, at times, easy to evaluate, often their outcomes are more unclear and ambiguous. Even for professional coaches or U.S. presidents, where there are clear starts and ends to their administrations due to contracts and term limits, the influences of leaders and their administrations may persist long after their contracts or terms end. For example, should the Obama Administration be attributed all the blame for the U.S. financial crisis of 2008 and, conversely, should the Clinton Administration be assigned all the credit for the economic prosperity in the late 1990s? Clearly, prior administrations and their actions or inactions may play a role. Assessing leadership outcomes is complicated further by the fact that there will seldom be a consensus among different constituents, even for leadership episodes widely recognized as abject disasters or highly effective. As noted earlier, inevitably some individuals will fare poorly from largely constructive leadership episodes, while some will fare well from generally destructive episodes.

However, regardless of personal opinions, on balance, there will typically be a majority opinion on a leadership episode’s eventual outcomes: destructive, constructive, or somewhere in between. Our points are that leadership episodes have starts and ends (none last forever) and that determining whether specific episodes are destructive requires waiting to evaluate the totality of group outcomes associated with them once they have exhausted their course. Again, while we remain sensitive to the misfortunes of individuals who may be negatively affected by a given leadership process, our purpose is to highlight the complex realities of these processes in organizations. Having discussed the central features of our definition, below we discuss areas of inconsistency and gaps in our knowledge of leaders, followers, and environments in destructive leadership and provide insight into these issues using our broader approach.

Leaders and Destructive Leadership

As we have noted, flawed or toxic leaders are obviously central to any destructive leadership story. Indeed, in a unique series of historiometric studies that coded biographies of historical leaders into quantitative indices, Mumford (2006) and O’Connor et al. (1995) found that personalized leaders, who focus on enhancing their power regardless of the costs to others and the organization (House and Howell 1992), were associated with greater harm to social systems. Thus, it is perhaps not surprising that leader characteristics (e.g., narcissism, personalized power motives, charisma) and behaviors (e.g., coercion, theft) have received the overwhelming attention from writers (Padilla et al. 2007). Yet, the literature includes several areas of conceptual confusion regarding leaders and destructive leadership, including (a) intentions, (b) negative influence versus counterproductivity, and (c) active versus passive behaviors. We seek to bring new insight to these issues below. In so doing and building on other attempts to develop an “all-inclusive” view of destructive leadership (e.g., Einarsen et al. 2007), we note that these processes may involve various “bad” leader behaviors. As alluded to by ecological, systems, and institutional theories, what is equally critical are the roles of followers and environments in cocreating destructive leadership.

Intentionality and Volitional Behaviors

Leader intentions complicate discussions of destructive leadership given intent is difficult, if not impossible, to assess. While some writers exclude unintended acts from their definitions, others include cases of thoughtlessness, ignorance, and ineptitude. For instance, Einarsen et al. (2007) argue that intent matters little, while Krasikova et al. (2013) suggest destructive leadership involves leaders consciously or unconsciously choosing potentially harmful goals or methods of influence. It is interesting to consider why some leaders would intentionally engage in actions that might harm their groups. One possibility is that they cannot help it; they are simply “bad” people who intend to do “bad” things and cannot control their behavior. Another possibility is that some leaders may believe their actions will help preserve their control or perhaps make their groups safer or more prosperous. Whether due to uncontrollable impulses or self-delusion, some leaders may deliberately do things that others would regard as “bad” and that may, under the right environmental circumstances, leave their groups and organizations worse off than before.

In our view, leader intentions are relevant just as leader behaviors are, but inevitably the group’s welfare is harmed or it is not. “Bad” leader intentions do not guarantee “bad” leadership outcomes, just like “good” leader intentions do not guarantee “good” leadership outcomes. Under certain conditions, a well-intentioned leader unaware of environmental changes or else inept could produce disastrous results via a massive blunder. Further, intent invariably matters little given group and organizational outcomes reflect more than what leaders do. As noted earlier, organizational ecology predicts that under conditions of environmental uncertainty, there are severe constraints on leaders’ ability to reliably devise and enact changes that increase the chances of organizational success and survival in the face of competition. In this view, even well-intentioned leaders can be associated with organizational “destruction” in uncertain environments. Furthermore, effective checks and balances and competent, autonomous followers may nullify the effects of intentionally or unintentionally “bad” leader behaviors on the group’s performance. Finally, it is difficult to distinguish clearly between a mistake or act of incompetence and a deliberate act of malice; it is challenging enough to fathom why individuals do what they do, let alone reliably assess whether they do so with intent. Evaluating whether a leader’s goals and actions are purposely harmful also creates the potential for social desirability in self-report measures and invalid inferences about the inner states of leaders in observational data (Krasikova et al. 2013). We do not observe intentions, only behaviors and outcomes. Thus, we argue destructive leadership may result regardless of a leader’s intent.

Negative Influence versus Counterproductivity

Negative influence (e.g., coercion) and counterproductive work behavior (CWB) (e.g., theft) are often considered in the literature. Some suggest destructive leadership involves the former, but not the latter (Schyns and Schilling 2013). For example, Krasikova et al. (2013) suggest destructive leadership entails behaviors that overlap with other negative actions (e.g., CWB, hostility), but that are “embedded in the process of leading” (p. 3). They argue such behaviors include encouraging followers to pursue “destructive” goals and using “destructive” modes of influence. Yet, like other leader-centric views, this assumes destructive leadership has occurred because the leader enacts certain behaviors. However, if followers do not comply, it is again not clear who is being “led” or what is being “destroyed.” As Hollander (1993) noted, “Without followers, there are plainly no leaders or leadership” (p. 29). Also, what is “destructive” influence to one person may be innocuous or acceptable to another. Thus, we argue that these behaviors do not ensure destructive leadership has occurred given their effects depend on followers and the environment.

Moreover, despite qualitative differences between CWBs and negative influence tactics, similar to other writers (e.g., Einarsen et al. 2007; Craig and Kaiser 2012), we believe both can be involved in destructive leadership processes. That is, while leader CWBs do not reflect explicit forms of influence like coercion, they may implicitly influence followers to enact “bad” behaviors and pursue potentially harmful goals. Indeed, modeling ethical behavior, for example, is a key facet of transformational (Bass 1985) and ethical theories of leader behavior (Brown and Treviño 2006). When leaders are free to engage in CWB, they act as role models for deviant behavior and implicitly convey its acceptability (Treviño et al. 2006). Without adequate controls, leader CWBs can permeate an organization and place it on a path to destruction (Ashforth et al. 2008). Enron is an example, whereby Jeff Skilling’s mark-to-market accounting practices and the subversion of oversight controls created a culture of corruption and greed.

Active Versus Passive Behaviors

Some writers omit passive leader behaviors from their definitions, noting qualitative differences between these behaviors and active forms of toxicity (Schyns and Schilling 2013). Yet, like other writers (e.g., Einarsen et al. 2007), we believe an inclusive definition of destructive leadership should include passive behaviors, such as delaying decisions, showing a lack of action or initiative, or neglecting one’s duties as a leader. This is because a leader’s passivity, under the right circumstances, may contribute to destructive outcomes. Blockbuster’s lack of strategic planning and failure to adapt to online and kiosk-based movie rentals, for example, contributed to the closing of 3500 stores nationwide in the U.S., thousands of lost jobs, and the firm’s eventual bankruptcy in 2011 (de la Merced 2011). Further, leaders who stress exploitation of current products while overlooking the need to explore new markets and technologies are often associated with firms that experience stagnated innovation and sizeable business losses over time (Benner and Tushman 2003). As such, we argue that destructive leadership may entail active and passive forms of “bad” leader behavior, although they are not sufficient to ensure destructive results will occur.

Followers and Destructive Leadership

Followers are the principle defenders against dysfunctional leaders (Kelley 2008). Yet, certain people, by nature of their traits and relationships with leaders and institutions, may comply with or even actively participate in destructive leadership processes. However, with a few exceptions, the roles of followers in destructive leadership remain underexplored. Below, we highlight three notable gaps in our understanding of followers and destructive leadership, including (a) followers’ susceptibilities, (b) their cocreational roles and influences in destructive leadership episodes, and (c) their development over time.

Follower Susceptibilities

Previous research has tended to focus on the power of situations in shaping people’s compliance with unethical or questionable authorities. Milgram’s (1963) studies of obedience, for example, revealed the willingness of participants to abandon their moral inclinations not to harm others in the presence of authority figures, while the Stanford Prison Experiment (Haney et al. 1972) showed how obedience can be shaped by a legitimizing ideology and institutional support. Yet, little is known regarding why certain people, due to their personal histories and dispositions, comply with or collude in destructive leadership processes. Such knowledge is vital given people differ in their reactions to flawed leaders no matter how powerful the situation (Kelman and Hamilton 1989). As Berkowitz (1999) warned, failing to account for follower susceptibilities limits our understanding of genocide and other real-world cases of destructive leadership.

In an attempt to illuminate the influence of personal factors on followers’ unique vulnerabilities, Thoroughgood et al. (2012a) developed a typology of susceptible followers, expanding on Padilla et al.’s (2007) distinction between conformers and colluders. While the former are prone to obedience, the latter actively engage in a leader’s toxic or misguided mission. Based on previous research, Thoroughgood et al. (2012a) proposed three types of conformers (lost souls, authoritarians, and bystanders) and two types of colluders (opportunists and acolytes). Although a discussion of this taxonomy is outside our scope, it highlights a key point: there are different followers associated with destructive leadership episodes who become a part of these processes for different reasons. For example, those with low self-concept clarity, negative self-views, and strong needs for security, affiliation, and meaning (lost souls) may be vulnerable to personal identification with and exploitation by charismatic leaders (Galanter 1980; Howell and Shamir 2005). In contrast, those marked by authoritarianism and cognitive rigidity (authoritarians) may comply with flawed leaders due to hierarchical values that prescribe a duty to obey authorities unconditionally (Altemeyer 1981) and a preference for leaders who exemplify order and discipline (Chirumbolo 2002).

Based on these susceptibilities, a number of questions emerge. For example, are some followers more likely to take part in destructive leadership processes depending on the leader and environment? That is, do certain “lock and key” patterns exist whereby matches between certain followers and leaders, within certain contexts, tend to produce destructive leadership outcomes? For instance, Machiavellian followers may conspire with self-serving leaders in loosely regulated corporations, where desires for wealth and power are more easily achieved. In contrast, authoritarian followers tend to form “toxic unions” with socially dominant leaders (e.g., Son Hing et al. 2007), perhaps especially in military settings where obedience is stressed. These questions are suggestive of only but a few of the many potential topics for future study.

Susceptible Follower Roles and Cocreational Influences

Once followers become a part of destructive leadership processes, as systems and institutional theories would suggest, their actions or inactions influence these processes and their outcomes. Yet, few analyses recognize the cocreational roles of followers in destructive leadership episodes. Most frame followers as passive victims, rather than key contributors to dysfunctional leader–follower relationships, conducive environments, and destructive outcomes. For example, the closest body of research related to followers’ roles in dysfunctional leadership situations comes from studies on victim precipitation, which suggest irritating aspects of followers (e.g., negative affect, low performance) prompt leaders’ hostility (Tepper et al. 2006, 2011) and use of coercion to induce follower compliance (Krasikova et al. 2013).

Such research reflects a step toward fuller appreciation of followers, yet it does not capture the complexities surrounding their unique roles in and contributions to destructive leadership episodes. Uhl-Bien et al. (2014) proposed that an understanding of followers in leadership processes requires role-based and constructionist perspectives. A role-based approach focuses on how individuals construe and enact their follower roles and the effects of their role behavior. A constructionist approach considers how following behaviors work in combination with leading behaviors to construct leadership processes and their outcomes. Thus, from this perspective, leadership processes require leading behaviors, but also following behaviors—otherwise leadership cannot take place (Hollander 1993; Uhl-Bien et al. 2014).

Extending these approaches to destructive leadership, key questions emerge regarding followers within these processes. From a role-based approach, how might conformers and colluders construe and enact their follower roles differently, producing unique effects on leaders, leader–follower relations, and the group? With respect to conformers, lost souls might construe their follower roles in terms of loyalty, obedience, and dependence on leaders for self-affirmation, leading to subservience, emulation of leaders, and attempts to garner their approval. As such, they may overempower self-serving leaders by fueling their needs for power, causing their leaders to forgo any ethical restraints on their use of power (Howell and Shamir 2005). In terms of colluders, opportunists, who follow dysfunctional leaders for personal gain (see Thoroughgood et al. 2012b), may define their follower roles in transactional terms. In exchange for financial or political outcomes, they may act as “yes” men, engaging in flattery, withholding criticism, and using manipulation to further the leader’s goals. This, in turn, may promote hubris in their leaders.

From a constructionist approach, the broader question that emerges is how follower conforming and colluding behaviors interact over time with actions or inactions of certain leaders (whether unethical, incompetent, passive, or otherwise flawed) within certain contexts to create destructive group outcomes? Consistent with Uhl-Bien et al. (2014), we argue that destructive leadership processes require combined acts of dysfunctional leading and following, which are left unchecked in the environment. Conformers and colluders must “grant” flawed leaders influence and “claim” their follower roles in these processes (Derue and Ashford 2010). Thus, when conformers (through dependence, blind obedience, or passivity) and colluders (via complicity or collaboration) permit such leaders influence and allow them to pursue their goals, they fail to fulfill their vital role as checks on their leaders’ power. Such following behaviors shape leader–follower dynamics marked by a lack of healthy scrutiny and centralized power in the leader.

In sum, the unique follower roles and following behaviors of susceptible individuals, as well as their interactions with flawed leaders, occupy a more central role in destructive leadership episodes than they have been granted to date. Consistent with systems theories, leaders and followers operate within the constraints they impose on one another and jointly shape the relationships that influence leadership outcomes (Howell and Shamir 2005). As such, more inclusive views on destructive leadership hinge on a greater appreciation for the cocreational roles that susceptible followers play within these processes.

Susceptible Follower Change and Development

Followers, like leaders, undergo cognitive changes as a result of the environments they inhabit. For example, social identity models (e.g., Hogg 2001) suggest leader–follower relations are a function of social-cognitive processes that cause individuals to redefine themselves based on distinct features of an in-group, leading to changes in their attitudes, motivations, and behaviors over time. With respect to destructive leadership processes, similar changes in followers may take place over time. For instance, Castro’s Pioneros, Mao’s Red Guard, and the Hitler Youth all underscore the transformative potential of vulnerable followers who internalize a leader’s toxic vision. Staub (1989) further noted how groups and societies marked by genocide travel along a “psychological continuum of destruction.” Group members experience psychological shifts whereby their motivations and inhibitions regarding harming out-group members change as they learn by doing and adopt attitudes that justify their behaviors. Some followers may develop a fanaticism for their group’s ideology—as shocking as it may be. Further, bystanders may undergo parallel shifts whereby they begin to devalue and distance themselves from victims, even becoming active contributors to the group in some cases. Similar processes may unfold in businesses, political institutions, and other organizations, whereby unethical behavior becomes entrenched, justified, and, in turn, results in the socialization of newcomers into unethical conduct (Ashforth and Anand 2003).

Recognition of followers’ developmental trajectories leads to many questions. For example, how might followers’ motivations for conforming and colluding shift as a result of exposure to their leaders, the dynamics among group members, and the context? As alluded to earlier, those with a malleable self-concept, negative self-views, and strong needs for group affiliation and a sense of purpose (lost souls) may become active colluders as they internalize a leader’s values. Followers’ motives for conforming or colluding may also intensify over time. For instance, an entrenched culture of greed may augment an opportunistic colluder’s desires for wealth and power. In sum, it is important to consider the dynamic nature of susceptible followers, recognizing that their motivations for following may change over time.

Environments and Destructive Leadership

Environments include the contexts, circumstances and conditions in which leaders and followers interact over time. Environments involve three categories of factors that may either promote or constrain dysfunctional leader–follower relations: institutional, macro-environmental, and cultural. Below, we list some of the factors that fall under these broad categories. In so doing, we draw on the fields of political science, sociology, and institutional economics, which focus on legal, economic, and governmental institutions and their effects on behavior. Understanding these factors offers insight into how conducive environments develop, become institutionalized, and shape dysfunctional forms of leading and following.

Institutional Elements

Institutional elements include major bodies that define, influence, and prescribe certain practices and behaviors, including legal, political, government, and collective bodies and institutions. Institutions can be efficient and functional at one extreme or inefficient and corrupt at the other. When institutional structures and processes are weak or absent, destructive leadership outcomes are more likely to occur.

Lack of Checks and Balances

While leaders require discretion to do their jobs, unconstrained authority provides opportunities for leaders and their followers to engage in toxic or ill-advise behavior. A lack of scrutiny is associated with weak institutional systems, including a lack of checks and balances on power. Checks and balances can be internal and external. The most important internal check is an organization itself: organizations that establish controls on their operations are able to more effectively regulate unethical or misguided leader–follower interactions. Internal checks include governing boards and other organizational processes and procedures—including fraud control systems, interdepartmental crosschecking, and ethics codes, committees, and communication networks. Indeed, research suggests fraud is greater in companies with weak governing boards than in those with stronger, more independent ones. For example, organizations lacking formal separation of the CEO and chair of the board positions and possessing fewer outside board members and weaker oversight controls demonstrate greater fraud (Beasley 1996; Beasley et al. 2000). Further, a lack of sufficient counsel by boards may also result in strategic mistakes on the part of leaders (Schnatter 2008). Other internal controls, such as ethics policies and monitoring systems, are also critical to curbing dysfunctional leader–follower relations (Treviño et al. 2006). Without these control mechanisms, these relationships can gain institutional momentum. For example, Ashforth and Anand (2003) discussed how an initial corrupt action can become entrenched in organizations, leading to selfish ideologies that excuse corruption. Socialization conveys the acceptance, or even desirability, of such behaviors to newcomers, allowing corruption to surpass individual tenures.

Examples of external checks include the media, the expert field, and government agencies. The media may function as a watchdog of leaders and followers through news coverage. In cases involving unseemly activity, the involved parties are typically forced to change their practices shortly after media scrutiny or stripped of power altogether. While leaders in other nations such as Russia and China engage in similar activities, media control in those two countries might prevent comparable disclosures. Thus, despite the media’s vital role in exposing malfeasance, the extent to which it is able to curb destructive leadership outcomes tends to vary based on its independence and vibrancy across cultures and nations.

Experts provide checks and balances by adding credibility to media reports, offering facts, and increasing objectivity. However, a lack of unbiased experts encourages unhealthy environments where toxic leader–follower relations remain unexamined. After the 2010 Deepwater Horizon oil rig explosion in the Gulf of Mexico, for example, journalists interviewed environmental and engineering experts who lent credibility to claims of negligence by BP. The constant media stories focused attention on other firms that contributed to and covered up facts regarding the spill. Three years later, Halliburton admitted to destroying evidence related to testing of the rig and had to pay a $200,000 fine, a voluntary $55 million donation to the Fisheries and Wildlife Foundation, and undergo 3 years of probation (Krauss 2013).

Finally, government agencies (e.g., the SEC, FTC, etc.) play an added role in regulating business activities of U.S. firms. Yet, inadequate funding and the fractured nature of the U.S. financial regulatory system often hinder the effectiveness of such agencies (Hitt 2009). Unlike other G10 nations (a group of ten industrialized countries that meet annually to consult and collaborate on international financial matters), which possess a central regulatory body, the U.S. system needs greater coordination among its many agencies, creating chances for toxic or misguided leader–follower activities to remain undetected until they result in destructive consequences. The problem is exacerbated by the lack of well-funded and competent regulatory agencies and rapid technological innovation. We discuss the latter problem below.

Macro-environmental Elements

Macroenvironmental factors represent broader social, economic, and technological forces that reduce organizational transparency, decrease checks and balances, and heighten followers’ dependence on leaders. The paradox is that some situations may require greater authority be allocated to leaders to promote quick, efficient responses to environmental demands. Yet, issues arise when certain industry features (e.g., complexity, instability) are paired with susceptible followers and flawed or toxic leaders.

Instability

Leaders can increase their power during times of crisis and instability when needs for clarity and order demand a leader with the perceived ability to resolve the crisis (Conger and Kanungo 1987). Under such conditions, leaders tend to be granted greater authority given instability requires swift action and often unilateral decisions (Janis and Mann 1977). Such was the case following the 9/11 attacks when U.S. President George Bush felt empowered to enact controversial antiterrorism provisions under the U.S. Patriot Act. Further, instability is related to perceptions of threat, which increase acceptance and support for autocratic leaders (Cohen et al. 2004) and foster centralized decision-making in firms (Keats and Hitt 1988). Threat perceptions can stem from internal (e.g., massive layoffs) or external stressors (e.g., market volatility). They can also range from feelings of mistreatment (e.g., Germany after Versailles) to desperate economic conditions (e.g., Mugabe’s Zimbabwe) to insolvency in a beleaguered corporation.

Complexity and Dynamism

Environmental complexity and dynamism may also reduce checks and balances on leaders’ power. Companies residing in industries characterized by rapid technological advancements and changing markets are harder to police given regulators must constantly adapt to and monitor complex networks of transactions and differentiated roles and tasks (Baucus and Near 1991). The interconnectedness of economies in the form of global firms that move across national borders with ease, avoiding local taxes and regulations in the process, further diminishes oversight and extends the potential reach of destructive leadership processes. For example, in addition to top executives’ effective lobbying of congressional leaders for favorable laws, Enron operated at a time when U.S. lawmakers generally favored deregulation of the energy industry—conditions that allowed leaders to conceal long-term contracts and complex financial trades (McLean and Elkind 2003). Regulatory accounting practices permitted Enron to bring more opaqueness to already complex and difficult-to-regulate financial tools.

Cultural and Societal Elements

Culture comprises the attitudes, beliefs, and values of a group, organization, region, or country and consists of multiple dimensions (Hofstede et al. 1990). Styles of leading in a U.S.-headquartered firm might be relatively egalitarian, reflecting the national and company culture, while management of an Indonesian subsidiary of the same firm may have a more authoritarian style, reflecting that nation’s traditions. Leader–follower relations can be greatly shaped by the culture in which they reside. Culture may produce additive effects that increase the conduciveness of environments to destructive leadership outcomes. Based on Hofstede’s (1984) work, unhealthy environments are likely to exist in cultures and organizations high on power distance, collectivism, and uncertainty avoidance. Power distance leads to tolerance of unequal power differentials and less dissent (Javidan et al. 2006). In collectivistic societies, leaders hold greater potential influence given collectivists tend to prefer strong leaders who bring people together and create group cohesion (Luthans et al. 1998). Finally, uncertainty avoidance leads to greater reliance on leaders for clarity and security; for example, dictators often exploit these needs by providing rules, regulations, and rituals that offer easy solutions to complex problems (Heifetz 1994). Together, these factors reduce scrutiny and undermine the independence and empowerment of potential dissenters.

Strategic Approaches to Studying Destructive Leadership Processes

To examine destructive leadership in a more holistic fashion, we recommend several approaches. First, following others (e.g., Avolio et al. 2009), we encourage more inductive investigations utilizing qualitative analysis. Qualitative studies focus on human interactions as they unfold in natural settings and can account for temporal changes in leadership processes over time. Interviews, case studies, and other qualitative methods are rare in leadership research, yet they provide many advantages, such as the ability to expose the rich inner workings of complex social-organizational processes (Silverman 2010). For example, in a case analysis of Bristol Royal Infirmary, a hospital that was subject to the longest medical inquiry in UK history, Fraher (2014) adopted a systems approach to examine how certain leader, follower, and environmental factors worked in combination, over 7 years from the pediatric cardiac surgical program’s start to its conclusion, to produce destructive outcomes, such as the deaths of dozens of babies undergoing surgery. Fields such as anthropology, sociology, and history all use ethnographic methods, which provide a more complete understanding of social systems via close analysis of human interactions in natural environments. Leadership episodes entail complex patterns of interactions among leaders and followers, followers and other followers, and leaders and leaders, all within specific contexts. Qualitative studies help unravel such patterns, detect themes and explanatory variables, and produce new theory. Thus, they often spur shifts in the way researchers think about and approach research questions.

Second, leadership scholars have showed a growing interest in historiometric analysis, a unique procedure permitting access to data not attainable using traditional surveys (e.g., Hunter et al. 2011; Yammarino et al. 2013). Historiometry relies on the coding of qualitative data obtained from verifiable historical sources into quantitative indices, which are then analyzed using traditional statistical analysis. With respect to leadership, data are most often derived from academic biographies of historical leaders. While not without limitations (see Ligon et al. 2012, Shamir 2011), historiometry permits tracking of various psychological, behavioral, and environmental factors that shape leadership processes. Mumford (2006), for example, examined the years spanning 120 historical leaders’ rise to, height of, and fall from power. As such, historiometric analysis can be used to compare factors related to leaders, followers, and environments over time and how they combine in unique ways to impact group and organizational outcomes.

Third, it is possible that well-planned longitudinal analyses could be employed to shed light on the ways in which leaders, followers, and environments interact over time to shape destructive outcomes. Yet, given the likely difficulties of accessing longitudinal survey data across lengthy leadership episodes (e.g., a CEO’s tenure), future investigations might target shorter, yet meaningful, periods—for example, the entire life of a temporary project team. Such studies might use meaningful start (e.g., appointment of a team’s leader) and end points (e.g., team disbandment). Although such analyses may lack the breadth and depth of qualitative and historical studies, they provide the advantage of greater quantitative rigor.

Conclusion

Destructive leadership processes reflect complex mosaics that cannot be understood by focusing primarily on leaders. By defining destructive leadership strictly in terms of leader behaviors, leader-centric views implicitly assume that “bad” leader behaviors are sufficient to create “bad” leadership outcomes, despite the reality that it is difficult to link these behaviors unequivocally with destructive outcomes for all groups, organizations, and their constituents. This is because the effects of these behaviors depend on contingencies related to followers and the environment. Relatedly, leader-centric views mask the influences of followers and environments, despite the fact that such factors are vital parts of destructive leadership processes. Without more balanced perspectives, potential solutions are not easily apparent because the process is not explained as a whole. By transcending beyond leaders and toward organizational outcomes and the contributing roles of followers and environments over time, more effective solutions and preventative remedies are possible.