Toxic Leadership: Definition, Traits, Destructive Leadership, Burnout, and Workplace Effects
Updated: 3 days ago
Author: Ukrainian Psychological Hub · Published: September 25, 2026 · Editorial Policy
“Toxic leadership” is widely used to describe leadership that repeatedly harms employees, teams, or organizations. The phrase is useful because it captures a recognizable workplace problem, but the science is more precise than the popular label. Researchers study several overlapping constructs—including destructive leadership, abusive supervision, petty tyranny, exploitative leadership, despotic leadership, supervisor undermining, and other forms of harmful leadership—and these constructs are not interchangeable. A leader can be ineffective without being toxic, strict without being abusive, narcissistic in trait measures without having Narcissistic Personality Disorder, and powerful without using coercive power.
The strongest evidence therefore comes from asking exactly what behavior was measured, who rated it, what outcome was studied, and what research design was used. A large 2021 systematic review and meta-analysis integrated 418 samples involving 123,511 working adults and found a broad, well-established network of adverse associations around destructive leadership (Mackey et al., 2021). A 2024 meta-analysis restricted to longitudinal field studies found that destructive leadership predicted poorer employee attitudes and behaviors across time, while also showing that some relationships are reciprocal and that stronger causal claims require care (Li et al., 2024).
This article explains what toxic leadership means, what its commonly measured traits and behaviors are, how it differs from neighboring constructs, what research shows about burnout and workplace outcomes, why personality labels are not diagnoses, and how organizations can respond without turning “toxic boss” into an all-purpose explanation for every difficult workplace relationship.
Quick Answer: What Is Toxic Leadership?
Toxic leadership is a broad label for persistent patterns of leader behavior and leader–follower interaction that produce meaningful harm to followers, teams, or organizations. In research, the term is often operationalized as a multidimensional pattern rather than a single behavior. Andrew Schmidt’s influential Toxic Leadership Scale, developed in military and civilian samples, organized toxic leadership around five dimensions: abusive supervision, authoritarian leadership, narcissism, self-promotion, and unpredictability (Schmidt, 2008). Later research has developed additional measures and definitions, which is one reason toxic leadership should not be treated as a perfectly standardized scientific construct.
For stronger conceptual precision, organizational research often uses destructive leadership as an umbrella family. Einarsen, Aasland, and Skogstad defined destructive leadership behavior as systematic and repeated leader, supervisor, or manager behavior that violates legitimate organizational interests by undermining organizational goals, resources, or effectiveness and/or subordinates’ motivation, well-being, or job satisfaction (Einarsen et al., 2007). Abusive supervision is narrower: Tepper’s foundational construct centers on subordinates’ perceptions of sustained hostile verbal and nonverbal supervisory behavior, excluding physical contact (Tepper, 2000).
The practical implication is important. “Toxic leadership” can be a useful entry point, but a scientific assessment asks what is actually happening: humiliation, threats, arbitrary punishment, exploitation, coercive control, self-serving behavior, unpredictable anger, undermining, chronic neglect, or something else. Different patterns have different measures, mechanisms, and evidence bases.
Toxic Leadership Is a Label, Not a Clinical Diagnosis
Toxic leadership is an organizational and leadership concept. It is not a diagnosis in the DSM or ICD, and a “toxic leader” is not a clinical category. The same rule applies to many words used casually at work. Narcissistic traits can be studied dimensionally in personality research, while Narcissistic Personality Disorder is a clinical diagnosis involving a pervasive pattern assessed in a diagnostic context. The American Psychiatric Association explicitly distinguishes everyday uses of “narcissist” from Narcissistic Personality Disorder (American Psychiatric Association, 2024).
Leadership research gives another reason for restraint. A meta-analysis found that narcissism was positively related to leader emergence but had no overall linear relationship with observer-rated leadership effectiveness; the emergence effect was largely explained by extraversion, and self-ratings of effectiveness were more favorable than observer ratings (Grijalva et al., 2015). Narcissistic traits may therefore matter for how people gain leadership roles or behave in them, but observing arrogance, self-promotion, or low empathy does not establish a personality disorder.
Psychopathy requires the same conceptual discipline. Psychopathy is a clinical and empirical construct with specialized assessment traditions, not a synonym for ruthlessness, coldness, or an unpleasant boss (Hare & Neumann, 2008). Organizational studies sometimes measure psychopathic tendencies dimensionally. A meta-analysis of 92 independent samples found small associations with leadership emergence and effectiveness and a more substantial negative association with transformational leadership, while also concluding that popular concern about “corporate psychopaths” can outrun the evidence (Landay et al., 2019). Workplace behavior should therefore be described behaviorally unless a qualified clinician has made an actual diagnosis.
Toxic Leadership, Destructive Leadership, and Abusive Supervision Are Not the Same Thing
The terms overlap, but they sit at different levels of precision.
Destructive leadership is a broad scholarly family concerned with leadership processes or behaviors that damage followers and/or organizations. It can include subordinate-directed harm, organization-directed harm, or both. Krasikova, Green, and LeBreton’s theoretical review showed why constructs such as abusive supervision, petty tyranny, and pseudo-transformational leadership capture parts of the destructive-leadership domain without exhausting it (Krasikova et al., 2013).
Toxic leadership is also broad, but its definitions and measurement models are less standardized. A recent review concluded that the field still lacks a unified conceptualization and cohesive measurement structure, despite growing empirical attention (Akinyele & Chen, 2025). Schmidt’s five-dimensional model is influential, but other researchers have proposed different dimensions and sector-specific scales. That makes toxic leadership a meaningful research label whose exact operational definition must be reported rather than assumed.
Abusive supervision is narrower and has been studied much more intensively. Tepper’s construct is based on subordinate perceptions of sustained hostile verbal and nonverbal supervisory behavior. A 2017 meta-analysis integrated its relationships with justice, individual differences, leadership variables, work attitudes, well-being, and behavior (Mackey et al., 2017). A later systematic review of 490 independent samples argued that the field still faces conceptual and measurement problems, especially because studies sometimes blur actual leader behavior with followers’ evaluations of that behavior (Fischer et al., 2021).
Simple incompetence is different from all three. A manager can be disorganized, indecisive, technically weak, poor at delegation, or promoted beyond their capabilities without engaging in a sustained pattern of hostility, exploitation, coercion, or intentional organizational damage. Incompetence can still produce serious consequences, but naming the mechanism correctly matters because skill deficits, role mismatch, and destructive behavior call for different interventions.
Leadership Style, Leadership Theory, Traits, and Behavior Also Need Separation
A leadership style is a recurring pattern of behavior, such as participative, directive, transformational, transactional, servant, or laissez-faire behavior. A leadership theory is an explanatory framework about how leadership works, why a pattern should matter, and under what conditions. The Hub’s guides to leadership styles and leadership theories treat these as distinct categories for precisely this reason.
Traits describe relatively enduring individual differences. Behaviors describe what a leader actually does in a situation. A trait may make some behaviors more likely without determining them. Toxic leadership research sometimes combines traits, motives, and behaviors in one measurement model—for example, narcissism and self-promotion alongside abusive or authoritarian behavior—so readers should not assume that every “toxic leadership trait” is a stable personality trait in the technical personality-psychology sense.
Management and supervision are neighboring constructs again. Management includes formal functions such as planning, allocating resources, coordinating work, staffing, monitoring, and controlling processes. Leadership concerns influence, direction, and collective action, although real organizational roles often combine both. Supervision is the formal oversight of another person’s work. A supervisor may exercise leadership, and a leader may have no supervisory authority. The distinction is developed in Leadership vs Management and becomes especially important when interpreting the abusive-supervision literature.
Power, Authority, Status, Dominance, and Prestige Are Different From Toxicity
Toxic leadership is often discussed in the language of “power,” but power itself is neither a diagnosis nor a form of leadership. Power concerns asymmetric control over valued outcomes or the capacity to affect others. Authority is a socially or institutionally recognized right to direct action. Status is socially conferred rank, respect, or standing. Dominance and prestige are distinct routes to rank and influence: dominance relies more on threat and coercive capacity, while prestige rests more on valued expertise and freely granted deference.
A formal manager can possess authority without much prestige. A respected expert can have status and influence without formal authority. A coercive leader can produce compliance without genuine commitment. A charismatic or prestigious leader can attract voluntary followership while still making poor decisions. These distinctions are developed in the Hub’s bridge article on power and leadership.
The distinction also prevents moral shortcuts. Power does not automatically equal abuse, and high standards do not automatically equal coercion. The relevant question is how influence is exercised. Coercive power is influence based on the capacity to punish noncompliance. Power and control separates control over other people from personal control and autonomy. Toxic leadership can involve coercive or controlling mechanisms, but those constructs should remain analytically separate.
What Traits and Behaviors Are Commonly Included in Toxic Leadership?
There is no single universally accepted toxic-leader profile. The most useful way to answer the “traits” question is to describe dimensions that have actually been operationalized and then separate them from neighboring constructs.
Abusive supervision
In Schmidt’s model, abusive supervision captures demeaning, belittling, or hostile treatment of subordinates. In the broader literature, abusive supervision has its own distinct measurement tradition derived from Tepper’s work. Examples can include ridicule, public humiliation, derogatory comments, hostile nonverbal behavior, or reminders of past failures used in a demeaning way. The construct is sustained rather than a single disagreement, and the foundational definition excludes physical contact.
This does not mean every blunt conversation is abusive supervision. Corrective feedback can be direct, performance standards can be high, and disciplinary action can be legitimate. The research construct concerns hostility and sustained mistreatment, not the mere fact that a subordinate disliked a decision.
Authoritarian control
Authoritarian leadership centralizes decision authority and restricts subordinate autonomy. In toxic-leadership measures, the concern is not simply that a leader makes decisions quickly or uses directive behavior during an emergency. The toxic pattern appears when centralized control becomes rigid, suppresses initiative, prevents reasonable participation, and operates alongside hostility, self-interest, unpredictability, or other destructive elements.
Context matters. Highly directive behavior can be functional when time is extremely limited, roles are standardized, or safety requires immediate coordination. The same behavior may become harmful when it is chronically unnecessary, arbitrary, or used to preserve personal dominance.
Narcissistic traits
Schmidt’s scale includes narcissism as a dimension, and destructive-leadership theories often discuss self-aggrandizement, entitlement, personalized use of power, and excessive self-focus. Yet “narcissistic leadership” should not become a diagnostic shortcut. Narcissism is measured in several ways, leadership outcomes vary by criterion, and observer-rated effectiveness does not show a simple linear relationship with narcissism (Grijalva et al., 2015).
In a toxic-leadership context, the behaviorally relevant questions are more concrete: Does the leader take disproportionate credit? React aggressively to criticism? Sacrifice team interests to protect status? Demand admiration? Devalue expertise that threatens their position? Those observations can be documented without diagnosing a personality disorder.
Self-promotion and self-serving leadership
Self-promotion refers to behavior aimed at advancing the leader’s image or career, especially when credit, blame, or resources are manipulated at others’ expense. A related research construct is exploitative leadership, defined around excessive leader self-interest and the exploitation of followers. Schmid, Pircher Verdorfer, and Peus developed and validated an exploitative-leadership measure across six studies involving 1,722 participants (Schmid et al., 2019).
Career ambition itself is not toxic. The distinction is whether advancement depends on appropriation, deception, instrumental treatment of followers, withholding opportunities, or other behavior that systematically transfers costs to others.
Unpredictability
Unpredictability refers to unstable, mood-dependent, or inconsistent leader behavior that makes the interpersonal environment difficult to anticipate. Employees may not know whether the same action will be praised, ignored, or punished on different days. Unpredictability matters psychologically because predictability helps people allocate attention, regulate behavior, and understand contingencies. When consequences seem arbitrary, more cognitive and emotional resources can be consumed by monitoring the leader rather than doing the work.
A single stressful day is not evidence of a toxic pattern. The relevant pattern is recurrent and consequential inconsistency.
Petty tyranny, humiliation, and arbitrary punishment
Ashforth’s classic work on petty tyranny described behaviors such as arbitrariness, self-aggrandizement, belittling, lack of consideration, forcing conflict-resolution styles, discouraging initiative, and noncontingent punishment (Ashforth, 1994). This construct helped establish that destructive management involves active patterns of domination rather than merely an absence of positive leadership.
Passive and avoidant leadership
Some destructive-leadership frameworks also include passive or laissez-faire forms. This requires care because passivity is different from hostility. Chronic avoidance of decisions, failure to intervene in conflict, refusal to provide necessary support, or absence during high-stakes problems can harm teams even when the leader is not aggressive. The mechanism is neglect, ambiguity, or failure of coordination rather than direct abuse. A comprehensive model of destructive leadership can therefore include active and passive routes to harm without pretending that they are the same behavior.
The Toxic Triangle: Harm Is Usually a System, Not Just a Personality
Padilla, Hogan, and Kaiser proposed the “toxic triangle” as a framework in which destructive outcomes arise from the interaction of destructive leaders, susceptible followers, and conducive environments (Padilla et al., 2007). The enduring value of this model is its shift away from a purely leader-centric explanation.
The leader side includes motives, behavior, charisma, personalized power, narcissistic tendencies, and other characteristics that can support destructive influence. The follower side concerns conditions that make resistance or correction harder, such as dependency, unmet needs, fear, ambition, strong identification, or incentives to collaborate. The environmental side includes instability, threat, cultural expectations, and weak checks and balances.
“Susceptible follower” should never be read as a synonym for culpable victim. Employees can be structurally dependent on a supervisor for pay, visas, references, promotion, scheduling, access to information, or professional reputation. Retaliation risk can make silence rational. A follower may recognize the problem clearly while lacking a safe way to challenge it. Organizational systems determine how costly resistance is.
The framework also explains why replacing one leader may not be enough. If the organization rewards short-term results while ignoring mistreatment, relies on opaque promotion systems, protects high-status performers from scrutiny, or has no credible reporting channel, another destructive leader can occupy the same niche. Leadership behavior is produced within an incentive and accountability architecture.
How Toxic Leadership Affects Employees
The evidence is strongest when toxic leadership is connected to the larger destructive-leadership and abusive-supervision literatures rather than relying on a single popular label.
Schyns and Schilling’s meta-analysis of 57 studies found destructive leadership negatively related to positive follower outcomes such as well-being, individual performance, and attitudes toward the leader, while positively related to outcomes such as turnover intentions, resistance, and counterproductive work behavior (Schyns & Schilling, 2013). The later Mackey et al. review substantially expanded the evidence base to 418 samples and 123,511 working adults (Mackey et al., 2021).
These findings do not mean every exposed employee will experience every outcome. Effects vary across people, jobs, leaders, measures, and contexts. They indicate population-level associations across a large literature.
Stress and psychological strain
Leaders shape job demands, access to resources, predictability, interpersonal threat, fairness, and the consequences of mistakes. Destructive leadership can therefore function as a chronic psychosocial stressor. A meta-analysis of leadership and stress integrated transformational leadership, leader–member exchange, and abusive supervision and found systematic relationships with follower stress and burnout (Harms et al., 2017).
The mechanism can be direct, as when a supervisor humiliates or threatens an employee, or indirect, as when uncertainty forces employees to remain constantly vigilant. It can also be relational: unfair treatment changes trust and social exchange. These mechanisms overlap but should not be treated as identical.
Burnout and emotional exhaustion
Burnout is more specific than feeling tired after a difficult week. The World Health Organization describes burnout in ICD-11 as an occupational phenomenon resulting from chronic workplace stress that has not been successfully managed, characterized by exhaustion, increased mental distance or cynicism toward the job, and reduced professional efficacy (World Health Organization). WHO does not classify burnout itself as a medical condition.
Destructive leadership is one possible workplace stressor among many. Workload, staffing, role conflict, low control, emotional labor, insecurity, harassment, organizational injustice, and resource scarcity can also contribute. It is therefore scientifically weak to infer “my leader caused my burnout” from temporal coincidence alone.
Still, the longitudinal evidence is meaningful. Li and colleagues’ meta-analysis of 82 articles covering 104 prospective or longitudinal studies and 30,314 employees found destructive leadership positively associated across time with burnout, emotional exhaustion, and turnover intention. In their pooled estimates, the longitudinal association was ρ = .296 for burnout, .270 for emotional exhaustion, and .354 for turnover intention (Li et al., 2024). These are associations across studies, not deterministic effects for an individual employee.
Job satisfaction, commitment, and engagement
Repeated hostility or arbitrariness can change how employees evaluate not only the leader but the job and organization. Longitudinal evidence links destructive leadership with lower job satisfaction, work commitment, and engagement. Li et al. reported a pooled longitudinal association of ρ = −.275 with job satisfaction and −.290 with work commitment, with the engagement estimate based on fewer studies (Li et al., 2024).
This distinction matters because an employee can remain highly competent while becoming psychologically detached from the organization. Reduced engagement is not equivalent to reduced ability.
Turnover intentions and leaving
Tepper’s foundational study found that employees who perceived more abusive supervision were more likely to quit; among those who remained, abusive supervision was associated with lower job and life satisfaction, lower affective and normative commitment, greater work–family conflict, and greater psychological distress (Tepper, 2000). Later meta-analyses support a robust relationship between destructive leadership and turnover intentions.
Intentions are not identical to actual turnover. Labor-market constraints, financial obligations, immigration status, professional specialization, family circumstances, and local opportunities can keep people in harmful environments even when they want to leave.
Performance, creativity, and citizenship behavior
The popular claim that toxic leadership “destroys performance” is directionally plausible but too simple. Performance effects are generally smaller and more heterogeneous than effects on attitudes toward the leader or turnover intentions. In the longitudinal meta-analysis, destructive leadership was negatively associated with job performance, creativity, organizational citizenship behavior, and voice, but the magnitudes varied and publication-bias adjustments reduced some estimates (Li et al., 2024).
This pattern makes psychological sense. A person can continue meeting formal performance requirements while trust, discretionary effort, creativity, or willingness to help deteriorates. Short-term output can therefore coexist with long-term social and organizational damage.
Silence, avoidance, resistance, and counterproductive work behavior
Employees do not respond to destructive leadership in only one way. Some withdraw, some remain silent, some resist, some avoid the supervisor, and some retaliate. Longitudinal meta-analytic evidence found positive associations with avoidance, resistance, silence, and counterproductive work behavior (Li et al., 2024).
These outcomes should not be moralized into a simple story in which “toxic leaders create toxic employees.” The same meta-analysis found evidence that negative employee behavior can also predict later destructive leadership. Leader and follower behavior can become reciprocal. Organizational conflict can spiral. That is precisely why longitudinal models are more informative than a one-time survey.
How Toxic Leadership Affects Teams and Organizations
Leadership operates through more than individual stress. A leader also shapes information flow, norms, conflict, incentives, and whether people believe it is safe to report errors.
When employees expect humiliation or retaliation, information can become strategically filtered. People may withhold bad news, avoid admitting uncertainty, reduce voice, or communicate defensively. This can make senior leaders incorrectly believe that a unit is functioning well because visible dissent has disappeared. Silence can therefore be an outcome of power relations rather than evidence of agreement.
Destructive leadership can also alter peer behavior. Social learning models predict that employees observe which conduct is rewarded or tolerated. If status is gained through intimidation and complaints are punished, aggressive norms can diffuse downward. Conversely, clear accountability can signal that high performance does not purchase permission to mistreat others.
Trust, psychological safety, engagement, inclusion, and job satisfaction remain distinct constructs. A team may trust a leader’s technical competence while feeling unsafe speaking up. An employee may be engaged with the work while distrusting management. Inclusion concerns belonging and participation, while psychological safety concerns the perceived interpersonal risk of speaking, asking, admitting, or challenging. Toxic leadership can affect several of these outcomes, but they should not be collapsed into a single “toxic culture” score.
Why a Toxic Leader Can Look Effective for a While
Toxic leadership and poor performance do not always appear at the same moment. A leader who centralizes decisions, pressures employees intensely, and uses fear may produce fast compliance or short-term output, especially when tasks are simple, monitoring is strong, and exit is difficult. That does not establish long-term effectiveness.
Leadership effectiveness is multidimensional. Relevant criteria can include unit performance, decision quality, retention, ethical conduct, adaptability, learning, follower well-being, safety, innovation, and long-term organizational viability. A leader can score well on one criterion while doing damage on another. Einarsen et al.’s model explicitly allows mixed forms—for example, behavior that appears pro-organization while being anti-subordinate (Einarsen et al., 2007).
This is also why charisma is not proof of effectiveness. Perceived charisma can increase attention, confidence, or voluntary identification, yet actual effectiveness depends on outcomes and context. Toxic-leadership analysis should therefore avoid the cinematic image of an obviously cruel, universally incompetent boss. Some destructive leaders are socially skilled, strategically competent, or rewarded by systems that value the outputs they produce.
Why “They Get Results” Is an Incomplete Defense
Organizations often tolerate harmful leadership when a person controls revenue, scarce expertise, client relationships, or a high-performing unit. Scientifically, this creates criterion contamination: “results” are defined too narrowly and the costs are displaced onto employees or future periods.
A more complete accounting includes turnover, replacement costs, sick leave, grievance handling, legal risk, lost knowledge, reduced voice, weakened collaboration, safety failures, reputation, and the opportunity cost of employees who stop contributing beyond the minimum. Some of these costs are delayed or difficult to measure, so the destructive leader’s short-term dashboard can look strong while the organization accumulates hidden liabilities.
The same logic applies to authority. Formal authority can make compliance easy to observe, but compliance is not equivalent to commitment, trust, or endorsement. The psychology of coercive power helps explain how punishment capacity can produce behavior without producing internal agreement.
What Causes Toxic or Destructive Leadership?
There is no single cause. The evidence points to multilevel antecedents involving leaders, followers, relationships, jobs, and organizations.
Individual differences can contribute without determining behavior
Narcissistic tendencies, antagonism, low empathy, self-interest, hostile attribution, impulsivity, and other traits may increase risk for some destructive behaviors in some contexts. But personality–behavior relationships are probabilistic. Traits do not excuse behavior, and they do not replace observation of behavior.
Popular Dark Triad language is especially prone to overreach. Narcissism, Machiavellianism, and psychopathy are distinct research constructs with different measurement traditions. Calling a difficult leader “Dark Triad” because they appear selfish or manipulative is not an assessment. The Hub’s dedicated articles on narcissistic leadership and Dark Triad leadership examine those narrower intents; this article uses personality evidence only to explain why trait labels should not substitute for behavioral analysis.
Stress and resource depletion can increase risk
Leaders also work under demands. High workload, role conflict, threat, fatigue, accountability pressure, and resource loss can impair self-regulation and increase reactive behavior. This can help explain why destructive behavior fluctuates without making it acceptable. An explanation identifies leverage points; it does not transfer responsibility to subordinates.
Incentives can reward destructiveness
If promotions depend only on short-term output, if employee turnover is invisible to senior leadership, if complaints damage the complainant’s career, or if star performers receive informal immunity, the organization creates selection pressure for destructive behavior. The leader does not need a pathological personality for the system to become harmful.
Weak checks and balances allow escalation
Padilla et al.’s toxic-triangle model emphasized the absence of institutional constraints as a conducive environmental factor (Padilla et al., 2007). In organizations, checks can include independent review, transparent promotion criteria, skip-level access, employee representation, audit trails, multiple data sources, and credible consequences for retaliation.
Power asymmetry changes the cost of disagreement
The more a leader controls schedules, evaluation, compensation, promotion, references, information, or continued employment, the more costly disagreement may become. That is a structural property of the relationship, not a personality trait. The Hub’s article on Power and Control explains why power over others should be distinguished from personal autonomy and control.
Why Followers Sometimes Stay
“Why don’t they just leave?” is a weak psychological question because exit has costs. Employees may have financial obligations, limited local alternatives, professional licensing constraints, health insurance dependencies, visa restrictions, caregiving responsibilities, pension considerations, or fear of reputational retaliation. They may also be attached to coworkers, clients, patients, a mission, or a career they do not want to surrender because of one leader.
Uncertainty also matters. Destructive leadership is often intermittent. Periods of hostility can alternate with praise, high performance, apology, or apparent normality. Employees may reasonably wait for more information, hope a restructuring will solve the problem, or believe formal processes will work.
Followers can also disagree about the same leader. Different employees receive different treatment, have different dependence relationships, or occupy different status positions. A leader may protect high performers who support them and punish dissenters. Therefore, “my team loves me” does not falsify reports of mistreatment, just as one negative report does not establish a universal pattern. Multi-source evidence matters.
Toxic Leadership and Workplace Bullying
Workplace bullying and destructive leadership overlap when a leader is the perpetrator, but they are not synonyms. Bullying research generally focuses on repeated exposure to negative acts where the target has difficulty defending themselves; destructive-leadership research focuses on the leadership or supervisory relationship and its harmful influence processes.
A leader can contribute to bullying without personally delivering every negative act—for example, by rewarding exclusion, ignoring repeated harassment, or creating a climate in which aggression is useful. Conversely, peer bullying can occur without a destructive leader. Clear case analysis should therefore identify who is acting, what they are doing, how often, and what power asymmetry is involved.
Toxic Leadership and Burnout: What the Evidence Actually Supports
The causal story is more complex than the slogan “toxic bosses cause burnout.”
First, burnout is a work-related occupational phenomenon, and leadership is one part of the work environment. Second, destructive leadership is associated with burnout and emotional exhaustion in meta-analytic research. Third, longitudinal studies strengthen the temporal case because leadership measured earlier predicts later outcomes. Fourth, observational longitudinal designs still do not create the control of a randomized experiment. Unmeasured job conditions, reverse effects, stable individual differences, and selection can contribute to the observed association.
Li et al.’s longitudinal meta-analysis is particularly useful because it tested reciprocal relationships. Destructive leadership showed lagged adverse relationships with multiple employee outcomes. Yet after controlling for autoregressive effects in cross-lagged models, destructive leadership did not significantly predict later negative employee behavior, whereas negative employee behavior predicted later destructive leadership (Li et al., 2024). This does not make destructive leadership harmless. It demonstrates that correlation, temporal prediction, reciprocal influence, and causal inference are different evidentiary claims.
For an individual employee, burnout should therefore be assessed in the context of the whole job. Leadership behavior may be a major stressor, one of several stressors, or a less important factor than workload, staffing, role conflict, or insecurity. The same person can also experience depression, anxiety, sleep problems, or medical conditions that require separate assessment. A leadership label cannot substitute for health evaluation.
How Strong Is the Scientific Evidence?
The evidence base is large, but its strengths differ by question.
Strongest: broad association with adverse workplace outcomes
Multiple meta-analyses and reviews converge on the finding that destructive leadership and abusive supervision are associated with poorer employee attitudes, well-being, and behavior and with greater turnover intention and counterproductive responses (Schyns & Schilling, 2013)|https://doi.org/10.1016/j.leaqua.2012.09.001]; [(Mackey et al., 2017)|https://doi.org/10.1177/0149206315573997]; [(Mackey et al., 2021). This is the most secure level of conclusion.
Strong: longitudinal prediction for several outcomes
The 2024 longitudinal meta-analysis synthesized 104 prospective or longitudinal studies and found lagged relationships with behavior, attitudes, burnout, emotional exhaustion, and turnover intention (Li et al., 2024). Longitudinal designs improve temporal inference, especially when baseline outcomes are modeled.
Moderate and context-dependent: specific mechanisms
Organizational justice, resource depletion, social exchange, identification, affect, and social learning all have plausible and empirically studied roles. But mediation evidence is often weaker than headlines imply because many mediation studies use cross-sectional or short time-lag designs. A statistical indirect effect is not automatically a demonstrated causal mechanism.
More limited: a single standardized toxic-leadership construct
“Toxic leadership” itself has less measurement consensus than abusive supervision or some other narrower constructs. Schmidt’s scale is influential, and newer sector-specific scales show active psychometric development, but researchers still disagree about which dimensions belong inside the construct (Akinyele & Chen, 2025). This is why an evidence-led article should not present a definitive checklist as though psychology had established one universal toxic-leader phenotype.
Limited for universal prevalence claims
Viral statistics about the percentage of workers who have a “toxic boss” are often based on proprietary surveys, varying definitions, or specific industries. Prevalence depends heavily on the construct, threshold, country, occupation, and measurement instrument. Estimates from nursing, military, hospitality, or one national sample should not be generalized to all workers.
Correlation, Prediction, Mediation, Longitudinal Evidence, and Causation
These terms describe different claims.
A correlation says two measured variables vary together. It does not establish direction.
Prediction means one variable statistically improves forecasts of another. Prediction can be useful without demonstrating a causal mechanism.
Mediation tests whether the statistical relationship between two variables is consistent with an indirect pathway through a third variable. Strong causal mediation requires temporal ordering and assumptions that many observational studies cannot fully satisfy.
Longitudinal evidence measures variables at more than one time. It can show temporal precedence and model change, but it remains vulnerable to confounding, attrition, measurement error, and incorrect time lags.
Causal evidence requires a design and assumptions that justify the claim that changing one variable would change another. In leadership research, randomized experiments with real sustained harmful leadership are often ethically impossible, so causal inference must be assembled from converging designs rather than from one perfect experiment.
This distinction is not academic decoration. It determines what organizations should do with the evidence. The case for taking repeated humiliation, intimidation, coercion, and destructive supervision seriously is already strong. The case for claiming a specific personality trait caused a particular employee’s burnout is much weaker.
Measuring Toxic Leadership
A “toxic leadership test” is not a diagnostic instrument. Research measures are designed to operationalize constructs for studies or structured organizational assessment.
Schmidt’s Toxic Leadership Scale emerged from work in military and civilian settings and includes five dimensions: abusive supervision, authoritarian leadership, narcissism, self-promotion, and unpredictability (Schmidt, 2008). Later researchers have proposed alternative toxic-leadership measures, including sector-specific instruments. That diversity is evidence of an evolving measurement field, not evidence that every checklist online measures the same thing.
Abusive-supervision measurement has a more established lineage, but it also has limitations. Fischer and colleagues’ systematic review argued that the dominant literature often conflates leader behavior with follower evaluation and relies heavily on single-source surveys (Fischer et al., 2021). A high-quality organizational assessment should therefore avoid treating one self-report scale as an objective verdict on a person.
Better assessment triangulates repeated behavioral reports, timing, context, multiple observers, relevant HR or operational records, turnover patterns, grievances, exit data, and opportunities for the leader to respond. It also separates behavior from impact: “raised voice in three meetings” is a behavioral report; “I felt unsafe” is an experience; “the leader is a narcissist” is an inference.
How Organizations Can Detect Destructive Leadership Earlier
The goal is to detect patterns before they become entrenched while minimizing false accusations and retaliation.
First, organizations can measure leader behavior from multiple directions. Upward feedback, peer feedback, direct observation, skip-level interviews, and climate data reveal different parts of the system. Anonymous surveys can increase reporting but should not be the only evidence used for high-stakes personnel decisions.
Second, organizations can watch for pattern convergence. One conflict may reflect a dyadic mismatch. Repeated independent reports involving humiliation, arbitrary punishment, retaliation, credit appropriation, or fear-based silence across time are more informative.
Third, outcome data can serve as signals rather than verdicts. Unusual turnover, internal transfers, absenteeism, grievances, safety events, or sudden drops in voice can trigger review. None of these proves toxic leadership on its own.
Fourth, organizations can examine whether reporting systems are credible. A nominal hotline is not a safeguard if complainants expect career damage, if investigations report directly to the leader involved, or if high-status employees are visibly exempt from rules.
Fifth, leader evaluation should include how results are achieved. If only output is rewarded, destructive externalities remain invisible. Multi-criterion evaluation makes retention, development, safety, ethics, and team functioning part of leadership effectiveness rather than optional “soft” outcomes.
What Can Organizations Do About Toxic Leadership?
Interventions should match the mechanism.
A skill deficit calls for development, coaching, role redesign, or clearer expectations. A conflict calls for structured resolution. A repeated abusive pattern may require formal investigation and consequences. A system that rewards coercion needs incentive and governance changes. A high-risk environment with weak checks and balances needs structural safeguards.
Training can help when leaders lack feedback skills, conflict-management skills, or awareness of behavioral impact. Training is less credible when the core problem is deliberate retaliation or exploitation and the organization has no willingness to enforce boundaries. Coaching should not become a substitute for accountability.
Selection can also improve, but personality screening is not a magic filter. Validated assessments can provide incremental information, yet destructive leadership emerges from person–situation interaction. Organizations that select “safe” personalities but reward abusive results can still manufacture destructive behavior.
Finally, senior leaders must examine protected performers. The most difficult cases are often not low performers but people who deliver valued outcomes while transferring costs to others. A system that excuses harm because “we cannot afford to lose them” teaches every observer what the organization actually values.
What Can an Employee Do Under a Toxic Leader?
The first task is to describe the problem accurately. Record specific behaviors, dates, contexts, witnesses, requests, and consequences when it is safe and lawful to do so. Concrete descriptions are more useful than labels such as “narcissist,” “psychopath,” or “toxic.”
Next, distinguish what can be changed locally from what requires structural support. A misunderstanding about priorities may respond to clarification. Repeated humiliation or retaliation usually requires a different route. Options can include a skip-level manager, HR, an ethics or compliance function, a union or professional body, an ombuds service, or an external adviser, depending on the workplace and jurisdiction.
Risk assessment matters. Direct confrontation is not always the safest or most effective strategy when the leader controls employment, immigration sponsorship, professional references, scheduling, or access to opportunities. Advice that assumes equal bargaining power can make the situation worse.
Health effects deserve their own response. Persistent sleep disturbance, anxiety, depressed mood, panic, physical symptoms, or functional impairment should be addressed as health concerns rather than waiting for the workplace dispute to resolve. Burnout is an occupational phenomenon, but a person can simultaneously have a diagnosable mental or physical condition that needs professional assessment.
Leaving can be a rational option, but it is not a moral obligation and is not equally available to everyone. The aim is to increase options and reduce exposure, not to blame an employee for remaining in a constrained situation.
When Difficult Leadership Is Not Toxic Leadership
Scientific precision protects both employees and leaders.
A leader can set demanding goals without being toxic. A supervisor can give negative feedback without being abusive. A manager can deny a promotion for legitimate reasons. A leader can make an unpopular restructuring decision. A crisis can require temporary centralization. A conflict can involve errors by both parties. A new manager can be clumsy while learning.
The boundary becomes more concerning when behavior is sustained, patterned, degrading, coercive, retaliatory, arbitrary, exploitative, or systematically harmful to followers or organizational functioning. Frequency, duration, intensity, power asymmetry, intent, impact, and organizational legitimacy all provide context.
Intent is not the only criterion. A leader may claim benevolent motives while repeatedly producing predictable harm. Conversely, harm alone does not prove destructive intent. This is another reason the Einarsen model is useful: it focuses on systematic and repeated behavior that undermines legitimate organizational interests or follower well-being, rather than requiring amateur mind-reading.
This boundary is an active research question rather than a settled piece of terminology. In a 2026 focal article, Ducheyne and De Fruyt argued for a sharper operational distinction between destructive or toxic leadership and incompetent leadership (Ducheyne & De Fruyt, 2026). Commentaries in the same journal issue immediately challenged aspects of that separation. Walters and Kreamer, for example, argued that incompetent leadership can itself be destructive and recommended focusing more directly on observable behavior and outcomes (Walters & Kreamer, 2026). The practical lesson is to document what the leader does, how often it occurs, what power is involved, and what consequences follow instead of treating “toxic” and “incompetent” as self-explanatory diagnoses.
Toxic Leadership Versus Leadership Failure
Leadership failure is the broader outcome category. A leader can fail because of poor strategy, hubris, groupthink, weak information processing, role mismatch, lack of expertise, bad incentives, inability to adapt, destructive behavior, or combinations of these. The Hub’s article Why Do Leaders Fail? owns that broader search intent.
Toxic leadership is one possible route to failure, but it can coexist with short-term success. Keeping the concepts separate prevents a circular definition in which “toxic” simply means “I dislike this leader” or “this leader’s unit eventually performed badly.”
Toxic Leadership Versus Leadership Effectiveness
Leadership effectiveness asks how well leadership produces relevant outcomes. Toxic leadership asks about harmful patterns and processes. The constructs can be negatively related without being exact opposites.
A leader may be interpersonally supportive but strategically ineffective. Another may deliver output while harming employees. Another may be both constructive and effective. A fourth may be destructive and ineffective. Effectiveness therefore needs multiple criteria.
This multidimensional view is central to Leadership Psychology, where leader emergence, traits, behaviors, followership, power, and effectiveness are treated as separate research questions rather than folded into one global judgment.
Toxic Leadership and Culture
“Toxic culture” is broader than toxic leadership. Culture concerns shared assumptions, norms, practices, incentives, and patterns of behavior that persist beyond one individual. A leader can shape culture, but culture can also constrain or amplify leaders.
If harmful conduct disappears when one leader leaves, the problem may have been concentrated. If the same pattern survives multiple leaders, the organization should examine structures, incentives, hiring, promotion, workload, governance, and reporting systems. The toxic-triangle perspective predicts exactly this interaction between people and environments.
Culture also affects interpretation. Directness, hierarchy, participation, emotional expression, and expectations of authority vary across occupations and societies. Cross-cultural variation does not make humiliation or retaliation harmless, but it means instruments developed in one setting should not be assumed to have identical meaning everywhere.
Common Myths About Toxic Leadership
“Toxic leaders are always narcissists.”
No. Narcissistic traits are one possible correlate or component in some models. Toxic or destructive leadership can emerge through many combinations of behavior, incentives, stress, power, and context. Narcissistic Personality Disorder is a clinical diagnosis and should not be inferred from workplace behavior.
“A toxic leader is just a bad manager.”
Not exactly. Bad management can reflect incompetence, unclear priorities, poor planning, weak delegation, or inadequate training. Toxic leadership usually refers to a repeated harmful pattern involving hostility, exploitation, domination, unpredictability, self-serving behavior, or comparable mechanisms. The categories can overlap.
“If the team performs well, the leader cannot be toxic.”
Short-term output is only one criterion. Destructive leadership can coexist with performance while increasing turnover, silence, exhaustion, or hidden risk. Effectiveness should be evaluated across multiple outcomes and time horizons.
“If employees stay, the situation cannot be that bad.”
Staying is shaped by opportunity, dependence, risk, identity, and economic constraints. Continued employment is weak evidence about the quality of leadership.
“One anonymous complaint proves toxic leadership.”
No. One report can identify a serious issue that deserves investigation, but a high-stakes conclusion should examine behavior, context, corroboration, pattern, and alternative explanations.
“A toxic-leadership score diagnoses the leader.”
No. Organizational scales measure responses to defined items. They do not diagnose a psychiatric disorder or establish an objective moral category.
FAQ
What is the simplest definition of toxic leadership?
Toxic leadership is a recurring pattern of leadership that meaningfully harms followers, teams, or organizational functioning. Because the term has multiple definitions, rigorous research should state the specific behaviors or measurement model being used.
What are the most common toxic leadership traits?
Commonly measured dimensions include abusive supervision, authoritarian control, narcissistic traits, self-promotion, and unpredictability in Schmidt’s Toxic Leadership Scale. Other research adds or emphasizes exploitation, petty tyranny, unethical behavior, passive neglect, or other destructive patterns. There is no single universally accepted trait profile.
Is toxic leadership the same as abusive supervision?
No. Toxic leadership is broader and less standardized. Abusive supervision is a specific research construct centered on subordinates’ perceptions of sustained hostile verbal and nonverbal supervisory behavior. Abusive supervision can be part of a toxic-leadership pattern, but the terms are not interchangeable.
Is toxic leadership the same as destructive leadership?
They overlap substantially, and some research uses toxic leadership within the broader destructive-leadership family. Destructive leadership has a more developed integrative literature and includes multiple forms of leader behavior that harm followers and/or organizations. Toxic leadership has competing definitions and measurement models.
Can toxic leadership cause burnout?
Destructive leadership is consistently associated with burnout and emotional exhaustion, including in longitudinal research. That supports destructive leadership as a meaningful workplace risk factor. For an individual person, however, burnout can reflect multiple job demands and resources, and observational evidence does not prove that one leader caused a specific case.
Is burnout a mental disorder?
WHO classifies burnout in ICD-11 as an occupational phenomenon, not a medical condition. Burnout can coexist with depression, anxiety, sleep disorders, or physical illness, which require their own assessment.
Is narcissism the cause of toxic leadership?
No single trait explains toxic leadership. Narcissistic traits are included in some models and can relate to leader emergence or behavior, but meta-analytic evidence does not show a simple linear relationship between narcissism and leadership effectiveness. Narcissistic Personality Disorder is a separate clinical diagnosis.
Are toxic leaders always ineffective?
No. A leader can deliver short-term output while producing employee harm, turnover, silence, or long-term organizational costs. Leadership effectiveness is multidimensional and should not be reduced to one performance metric.
Can a strict leader be non-toxic?
Yes. High standards, direct feedback, rapid decision-making, and accountability are not inherently toxic. The scientific concern is sustained harmful behavior such as humiliation, arbitrary punishment, coercive control, exploitation, retaliation, or comparable patterns.
Why do organizations keep toxic leaders?
Possible reasons include narrow performance incentives, dependence on scarce expertise or revenue, weak reporting systems, fear of disruption, status protection, poor governance, and delayed visibility of turnover or health costs. These are organizational explanations, not just personality explanations.
What is the difference between coercive power and toxic leadership?
Coercive power is a source of influence based on the capacity to punish noncompliance. Toxic leadership is a broader pattern of harmful leadership. A leader can possess coercive power without routinely abusing it, and toxic leadership can use mechanisms other than punishment.
What should organizations measure?
They should measure specific leader behaviors, employee experiences, relevant outcomes, and organizational context using multiple data sources. One global “toxicity” score is less informative than a well-designed assessment of repeated behaviors, frequency, power asymmetry, retaliation risk, turnover, voice, and well-being.
Related Articles
References
Akinyele, A. I., & Chen, Z. (2025). Dark clouds of leadership: Causes and consequences of toxic leadership. International Studies of Management & Organization, 55(4), 476–503. https://doi.org/10.1080/00208825.2024.2442185
American Psychiatric Association. (2024). What is narcissistic personality disorder? https://www.psychiatry.org/News-room/APA-Blogs/What-Is-Narcissistic-Personality-Disorder
Ashforth, B. (1994). Petty tyranny in organizations. Human Relations, 47(7), 755–778. https://doi.org/10.1177/001872679404700701
Ducheyne, D., & De Fruyt, F. (2026). Distinguishing destructive/toxic from incompetent leadership. Industrial and Organizational Psychology, 19(3), 245–258. https://doi.org/10.1017/iop.2026.10070
Einarsen, S., Aasland, M. S., & Skogstad, A. (2007). Destructive leadership behaviour: A definition and conceptual model. The Leadership Quarterly, 18(3), 207–216. https://doi.org/10.1016/j.leaqua.2007.03.002
Fischer, T., Tian, A. W., Lee, A., & Hughes, D. J. (2021). Abusive supervision: A systematic review and fundamental rethink. The Leadership Quarterly, 32(6), 101540. https://doi.org/10.1016/j.leaqua.2021.101540
Grijalva, E., Harms, P. D., Newman, D. A., Gaddis, B. H., & Fraley, R. C. (2015). Narcissism and leadership: A meta-analytic review of linear and nonlinear relationships. Personnel Psychology, 68(1), 1–47. https://doi.org/10.1111/peps.12072
Hare, R. D., & Neumann, C. S. (2008). Psychopathy as a clinical and empirical construct. Annual Review of Clinical Psychology, 4, 217–246. https://doi.org/10.1146/annurev.clinpsy.3.022806.091452
Harms, P. D., Credé, M., Tynan, M., Leon, M., & Jeung, W. (2017). Leadership and stress: A meta-analytic review. The Leadership Quarterly, 28(1), 178–194. https://doi.org/10.1016/j.leaqua.2016.10.006
Krasikova, D. V., Green, S. G., & LeBreton, J. M. (2013). Destructive leadership: A theoretical review, integration, and future research agenda. Journal of Management, 39(5), 1308–1338. https://doi.org/10.1177/0149206312471388
Landay, K., Harms, P. D., & Credé, M. (2019). Shall we serve the dark lords? A meta-analytic review of psychopathy and leadership. Journal of Applied Psychology, 104(1), 183–196. https://pubmed.ncbi.nlm.nih.gov/30321033/
Li, P., Yin, K., Shi, J., Damen, T. G. E., & Taris, T. W. (2024). Are bad leaders indeed bad for employees? A meta-analysis of longitudinal studies between destructive leadership and employee outcomes. Journal of Business Ethics, 191, 399–413. https://doi.org/10.1007/s10551-023-05449-2
Mackey, J. D., Frieder, R. E., Brees, J. R., & Martinko, M. J. (2017). Abusive supervision: A meta-analysis and empirical review. Journal of Management, 43(6), 1940–1965. https://doi.org/10.1177/0149206315573997
Mackey, J. D., Ellen, B. P., III, McAllister, C. P., & Alexander, K. C. (2021). The dark side of leadership: A systematic literature review and meta-analysis of destructive leadership research. Journal of Business Research, 132, 705–718. https://doi.org/10.1016/j.jbusres.2020.10.037
Montano, D., Reeske, A., Franke, F., & Hüffmeier, J. (2017). Leadership, followers’ mental health and job performance in organizations: A comprehensive meta-analysis from an occupational health perspective. Journal of Organizational Behavior, 38(3), 327–350. https://doi.org/10.1002/job.2124
Padilla, A., Hogan, R., & Kaiser, R. B. (2007). The toxic triangle: Destructive leaders, susceptible followers, and conducive environments. The Leadership Quarterly, 18(3), 176–194. https://doi.org/10.1016/j.leaqua.2007.03.001
Schmid, E. A., Pircher Verdorfer, A., & Peus, C. (2019). Shedding light on leaders’ self-interest: Theory and measurement of exploitative leadership. Journal of Management, 45(4), 1401–1433. https://doi.org/10.1177/0149206317707810
Schmidt, A. A. (2008). Development and validation of the Toxic Leadership Scale [Master’s thesis, University of Maryland]. DRUM. http://hdl.handle.net/1903/8176
Schyns, B., & Schilling, J. (2013). How bad are the effects of bad leaders? A meta-analysis of destructive leadership and its outcomes. The Leadership Quarterly, 24(1), 138–158. https://doi.org/10.1016/j.leaqua.2012.09.001
Tepper, B. J. (2000). Consequences of abusive supervision. Academy of Management Journal, 43(2), 178–190. https://doi.org/10.5465/1556375
Walters, Z. A., & Kreamer, L. (2026). A distinction with little difference: Incompetent leadership is destructive. Industrial and Organizational Psychology, 19(3), 271–274. https://doi.org/10.1017/iop.2026.10101
World Health Organization. (n.d.). Burn-out an occupational phenomenon. https://www.who.int/standards/classifications/frequently-asked-questions/burn-out-an-occupational-phenomenon