Research on leadership selection consistently finds that charisma, defined loosely as the capacity to project confidence, warmth, and vision in social interactions, is among the strongest predictors of advancement in organizations. It is also, when examined carefully against outcomes data, a relatively weak predictor of actual leadership effectiveness. The gap between these two findings is not a statistical anomaly. It is a structural feature of how organizations identify and advance leadership talent, and its consequences for organizational performance are substantial and underappreciated.

The research foundation for this observation is extensive. A meta-analysis of leadership emergence studies found that individuals perceived as charismatic were significantly more likely to be identified as leaders by their peers and superiors, and more likely to be selected for leadership roles, than individuals who demonstrated equivalent or superior analytical and interpersonal competence without the charismatic presentation. A separate body of research on leader effectiveness found that charisma, while positively correlated with follower satisfaction and perceived leader quality, showed weak and inconsistent correlations with objective measures of organizational performance: revenue growth, operational efficiency, employee retention, and strategic execution quality.

The gap between selection and performance is not random. It reflects a systematic feature of how organizations identify leadership potential and the contexts in which those identification processes operate. The primary contexts in which promotion decisions are made — interviews, presentations to senior leadership, performance reviews, participation in high-visibility meetings, social interactions with organizational decision-makers — are precisely the contexts in which charisma is most legible and most impressive. They are also contexts that are structurally ill-suited to revealing the skills that predict leadership effectiveness.

Analytical rigor is not visible in a presentation; it is visible in the quality of decisions made under uncertainty over time. Tolerance for conflict is not visible in a meeting with senior leadership; it is visible in how a manager handles disagreement within her team when no one senior is watching. Willingness to deliver unwelcome information is not visible in an interview; it is visible in how a leader communicates bad news to her own superiors when the personal cost of honesty is high. Capacity for sustained attention to unglamorous operational detail is not visible in any of the contexts where promotion decisions are typically made. The selection environment systematically surfaces the skills that charisma makes legible while obscuring the skills that organizational effectiveness requires.

The result is a selection mechanism that reliably identifies the people who are best at being evaluated for leadership, rather than the people who are best at leading. These are not the same population. The skills required to perform well in a leadership selection context — the capacity to present confidently, to project competence and warmth simultaneously, to make complex ideas sound simple and compelling, to read a room and calibrate communication accordingly — are genuinely valuable skills. They are simply not the primary skills that predict whether an organization will execute its strategy, retain its talent, manage its risks, or make good decisions under pressure. Organizations that optimize their selection processes for the former set of skills while believing they are selecting for the latter should not be surprised when the leaders they identify underperform against the expectations their selection created.

The mechanism is further distorted by what researchers have called the romance of leadership: the systematic attribution of organizational outcomes, both positive and negative, to the quality of individual leadership rather than to contextual, structural, and market factors. Leaders who inherit favorable conditions — a growing market, a strong competitive position, a well-functioning organizational structure built by their predecessors — and preside over good outcomes are evaluated as excellent. The positive outcomes are attributed to their vision and competence. Leaders who inherit difficult conditions and preside over poor outcomes are evaluated as inadequate. The negative outcomes are attributed to their failures rather than to the circumstances they inherited. The signal from this evaluation is almost entirely noise from the perspective of identifying what actually produced the outcomes. But it shapes subsequent selection criteria, compensation structures, and organizational narratives as if it were reliable information about individual leader quality.

The charismatic leader who presides over a favorable period and is credited with the organizational success that follows becomes the template against which future leaders are selected. The specific qualities that made her charismatic — the confidence, the vision, the compelling communication — are identified as the causal factors in the organization’s success and are sought in subsequent selection processes. Whether those qualities actually caused the success, or whether the success would have occurred under a range of leadership styles given the favorable conditions, is a question that organizational attribution processes are not designed to ask.

The structural interventions that reduce the influence of charisma on leadership selection are well documented, if inconsistently applied. Structured interviews with standardized behavioral questions reduce the advantage of unguided social performance. Work sample assessments that require candidates to demonstrate the actual skills of the role rather than the skills of the selection context improve predictive validity substantially. Blind evaluation of analytical work product separates the quality of thinking from the social performance that typically accompanies its presentation. Multi-rater assessments that include the perspectives of direct reports, peers, and internal stakeholders who have observed the candidate in conditions that resemble the actual demands of the role provide information that upward-facing performance in selection contexts systematically omits.

These interventions are not new. The evidence supporting them has been available for decades. The resistance to implementing them at scale is itself a phenomenon worth examining. Organizations whose senior leadership was selected through processes that rewarded charismatic performance have limited incentive to redesign those processes in ways that would have selected differently. The people who benefited from the existing selection criteria are, in most organizations, the people with the authority to change them. This is not a conspiracy. It is loss aversion and self-concept maintenance operating at the institutional level, producing predictable resistance to reforms that would, in expectation, improve organizational outcomes while threatening the implicit validation that the existing selection criteria provide to those who succeeded under them.

Laurence Peter published his principle in 1969. Fifty-three years later, the research literature continues to confirm it. Employees are routinely promoted to positions that require skills they do not have, based on demonstrated excellence in positions that required different skills. The result is organizations populated, at every level above entry, with people operating at or near their threshold of incompetence.

The original formulation was satirical in register but serious in substance. Peter and Hull were describing something that practitioners recognized immediately and that subsequent empirical research has confirmed repeatedly. A 2019 study published in the Quarterly Journal of Economics examined promotion patterns across 214 companies and found that the best-performing individual contributors were consistently more likely to be promoted into management than their less individually productive peers, and that this promotion pattern was negatively correlated with subsequent team performance. The best salespeople, promoted into sales management, produced teams that performed worse than the teams of managers who had been less exceptional as individual contributors. The finding was robust across industries and organizational types.

What the original formulation missed, and what subsequent behavioral economics research has clarified, is the mechanism by which this happens. It is not simply that organizations fail to assess the skills required for higher-level positions. It is that the skills most visible and most legible to the people making promotion decisions are precisely the skills that predict success in the role being vacated, not the role being filled.

A technically excellent individual contributor is visible as excellent along the dimensions that individual contribution makes salient: output quality, problem-solving speed, domain knowledge, and the capacity for sustained independent effort. What is not visible, because it is not required in that role, is the capacity for motivating others, tolerating ambiguity, delegating effectively, delivering unwelcome assessments to people one is responsible for developing, and managing upward with the combination of candor and political awareness that organizational effectiveness requires. These are the skills that predict managerial success. They are systematically invisible during the period when promotion decisions are being made.

The problem is compounded by the social dynamics of promotion decisions. The people making promotion choices have typically observed the candidate performing in the current role over an extended period. They have developed a concrete, evidence-based assessment of the candidate’s competence along the dimensions the current role requires. The assessment of readiness for the new role, by contrast, is necessarily speculative and abstract. Under cognitive load and time pressure, decision-makers default to the concrete over the abstract. The candidate who is demonstrably excellent now is promoted over the candidate who might be more suited to the new role, because demonstrated current excellence is legible in a way that projected future suitability is not.

This is further distorted by what researchers have called the halo effect: the tendency to attribute positive qualities broadly to people who have demonstrated excellence in specific domains. The exceptional engineer is assumed to have the interpersonal and organizational skills that management requires, not because there is evidence for this but because excellence in one domain generates a general presumption of excellence. The halo effect is well-documented in laboratory settings. Its organizational consequences are underappreciated.

The organizations that have most successfully disrupted this pattern share a structural feature: they have separated the criteria for promotion from the criteria for current role performance and built explicit assessments of the skills required for the destination role rather than the origin role. This is not complicated in principle. It requires only that organizations accept that excellence in one role is not evidence of readiness for a different role, and build processes that reflect that acceptance. Assessment centers, structured behavioral interviews focused on managerial competencies, trial periods in acting roles, and the formal separation of individual contributor and management career tracks are all interventions with empirical support.

Most organizations have not implemented these interventions at scale. Most still promote their best salespeople into sales management, their best engineers into engineering management, their best clinicians into clinical administration. The research on the consequences is unambiguous and has been for decades. The practice continues because the alternative requires organizations to tell their best performers that excellence in their current role is not sufficient qualification for the next one. That conversation is difficult. The organizational cost of avoiding it is substantially higher than the cost of having it, but that cost is diffuse and delayed in a way that makes it easy to discount. This is itself a form of organizational self-deception: the substitution of short-term social comfort for long-term organizational effectiveness.