Quiet exits, activist pressure, and a rolling debate over where and how people work have turned CEO succession into a cliff-edge moment. Boards still plan as if the future hinges on finding one heroic leader. The research points a different way: distribute authority, then rotate it with intent. The strongest evidence sits at the team level, and it consistently links shared leadership with better results. The question is how far that model can travel toward the top.
What the evidence actually says
Two independent meta-analyses synthesize decades of team research and reach a similar conclusion. Shared leadership, where influence is distributed across members rather than concentrated in one formal leader, is positively related to performance. D'Innocenzo, Mathieu and Kukenberger reviewed 50 effect sizes covering 3,198 teams and found a clear positive relationship, with stronger effects in real organizational settings than in classrooms or labs (2016). Importantly, how researchers measured sharing mattered. Studies that mapped actual influence networks, such as the density and decentralization of who leads whom, showed larger performance gains than studies that simply aggregated individual ratings. That points to a design lesson: the value comes from real patterns of distributed influence, not a title shuffle.
Wang, Waldman and Zhang, in a separate meta-analysis of 42 samples, reported an overall correlation of about .34 between shared leadership and team effectiveness (2014). They also found stronger effects when work is complex and when the shared behaviors are the so-called new-genre forms, such as transformational leadership, rather than only traditional task structuring. For knowledge work and cross-functional initiatives, that is not a small detail. The more a team faces ambiguity and interdependence, the more it benefits when leadership is a dynamic property of the work rather than a static role.
These are team-level findings, not CEO-level proof. They do not license glib claims that one can simply rotate the corner office and expect the same effect. They do, however, frame a hypothesis worth testing at the top: when the work is complex and interdependent, organizations perform better when decision rights and influence travel to the person best positioned to exercise them, then travel again as conditions change.
Can the model scale to CEOs?
Shared top roles are rare but not mythical. Feigen, Jenkins and Warendh examined 87 companies with co-CEOs among roughly 2,200 firms in the S&P 1200 and Russell 1000 between 1996 and 2020, and argue the model is more viable than its reputation suggests (2022). The authors are clear about the drawbacks. They note prominent failures, including at Chipotle Mexican Grill, SAP, and Research In Motion/BlackBerry, especially under stress. The rarity itself, fewer than 100 such cases in a large index set over nearly a quarter century, is a signal of caution.
Yet the comparison to the team evidence still holds a lesson. Where shared leadership worked best in teams, the pattern was real and visible in the network, not performative, and it was observed in field settings. Translated to the C-suite, co-leadership that divides domains with precision, sets a cadence for joint decisions, and makes the flow of authority explicit is closer to the higher-performing cases implied by network-based measures in D'Innocenzo, Mathieu and Kukenberger (2016). By contrast, a cosmetic co-CEO structure that obscures who decides what resembles the weaker aggregation-based patterns that meta-analysis found less predictive of performance.
Boards do not need to jump straight to permanent co-CEOs. Rotating leadership can start with time-bound mandates for mission-critical domains, for example elevating a product lead to steward a platform shift for two quarters while the operating chief runs core P&L. The point is not to erase accountability. It is to align the locus of leadership with the locus of expertise, then revisit it frequently as the problem shifts. That is the practical extension of Wang, Waldman and Zhang's finding that shared leadership pays off most when the work is complex and the influence behaviors are adaptive rather than purely directive (2014).
Designing rotation without chaos
The mechanics matter. The team meta-analyses reward actual influence patterns, not theory, which suggests several design moves are prudent interpretations rather than proven prescriptions.
-
Make decision rights explicit. Publish who decides, who must be consulted, and what the time horizon is for each rotating mandate. This is how you create the equivalent of a visible influence network, the context in which performance effects were strongest in D'Innocenzo, Mathieu and Kukenberger (2016).
-
Set a fixed rotation cadence with formal handoffs. Treat rotation as a governance process, not a personality contest. Written transition memos, joint reviews at the midpoint, and a closing retrospective keep influence moving while retaining continuity. The field settings in which shared leadership linked more strongly to performance point to the value of disciplined practice over ad hoc swaps (D'Innocenzo, Mathieu and Kukenberger, 2016).
-
Match rotation to complexity. Focus first on domains where interdependence and ambiguity are highest, such as platform migrations, new-market entries, or cross-functional cost programs. This aligns with the stronger effects for complex work reported by Wang, Waldman and Zhang (2014).
-
Preserve a single point of escalation. Even in co-CEO settings, crises punish ambiguity. Feigen, Jenkins and Warendh note that several dual-leadership arrangements faltered under stress, a reminder to predefine tie-breakers and crisis command (2022).
These steps do not claim causal proof at the CEO level. They translate what the evidence reliably shows in teams into a board-level experiment with guardrails.
A succession plan you can practice, not just draft
The practical appeal of rotating leadership is that it turns succession from a once-a-decade gamble into a continuous capability. By regularly shifting who leads critical work, boards see future CEOs under realistic pressure and teams build the muscle to follow rotating authority. The meta-analyses show that when leadership is shared in practice, not just in charts, performance improves in the places that most resemble executive work: complex, interdependent, and fluid (Wang, Waldman and Zhang, 2014; D'Innocenzo, Mathieu and Kukenberger, 2016). The co-CEO evidence is more tentative and mixed, but it establishes that shared top roles are feasible under the right conditions and that poor design invites failure (Feigen, Jenkins and Warendh, 2022).
The implication for boards is concrete. Pilot rotational mandates in the executive committee for one year. Predefine domains and decision rights. Publish the cadence. Add a crisis escalation rule. Then measure what matters: decision speed, strategic throughput, and followership quality. If the experiment yields crisper execution in complex work and stronger bench readiness, keep going. If it produces diffusion of responsibility, adjust the design or reconsider the scope.
Shared and rotating leadership is not a slogan. It is a choice to make influence visible, mobile, and accountable. The science at the team level is strong. The path to the corner office is a translation challenge, not a leap of faith.
Sources
- D'Innocenzo, Mathieu and Kukenberger (2016). A Meta-Analysis of Different Forms of Shared Leadership–Team Performance Relations, Journal of Management. Synthesizes 50 effect sizes across 3,198 teams, finding a positive shared-leadership–performance link that is stronger in field settings and when measured as real influence networks. https://journals.sagepub.com/doi/10.1177/0149206314525205
- Wang, Waldman and Zhang (2014). A meta-analysis of shared leadership and team effectiveness, Journal of Applied Psychology. Finds an overall positive relationship (rho ≈ .34), stronger in complex work and for new-genre leadership behaviors, supporting the case for distributed influence in knowledge-intensive contexts. https://pubmed.ncbi.nlm.nih.gov/24188392/
- Feigen, Jenkins and Warendh (2022). Is It Time to Consider Co-CEOs?, Harvard Business Review. Reviews 87 co-CEO companies, noting the arrangement is rare and mixed, with notable failures under stress, yet argues shared top leadership is more viable than assumed when well designed. https://hbr.org/2022/07/is-it-time-to-consider-co-ceos