The Succession Trap

The Puzzle in the Data

In the first half of 2026, two of the world's most authoritative leadership advisory firms released their annual readings of CEO succession — and the conclusions pointed in opposite directions.

Russell Reynolds Associates reported that boards at the world's largest public companies were holding on to their chief executives longer than at any point in the past decade. Average departing tenure had reached 9.0 years, up from 6.6, and total global CEO departures fell to 101 — the lowest first-half count in nine years. When these boards did hire, they increasingly demanded candidates who had already held a public-company CEO role: a record 23% globally and 34% in the S&P 500¹. The message was unambiguous. At the top of the market, boards wanted proven leaders and were in no hurry to replace the ones they had.

Weeks later, Spencer Stuart published its analysis of the broader S&P 1500 — and the picture inverted. The index recorded 168 new CEO appointments in 2025, the most since 2010, with 84% of those incoming leaders serving as enterprise CEO for the first time. Average tenure had fallen to 8.5 years, the lowest since 2019. Nearly 40% of departing CEOs left within their first five years. Technology, media and telecom transitions nearly doubled year over year².

One dataset says boards are demanding more experience than ever. The other says they are installing less of it than at any point in fifteen years. Both cannot describe the same market doing the same thing. But they can describe the same system failing in two different ways — and that is what the evidence, taken together, reveals.

Divergence, Not Contradiction

The contradiction resolves once you recognise that these reports are measuring different segments of the same ecosystem. The Russell Reynolds index tracks the world's largest public companies, where governance scrutiny, regulatory complexity and the sheer scale of the role create powerful incentives to retain known quantities. Spencer Stuart's analysis covers the broader S&P 1500 — a universe that includes mid-cap and small-cap companies where disruption runs hotter, strategic pivots come faster, and the pressure to install new leadership is harder to defer.

What appears contradictory is in fact a bifurcation. The large-cap market and the broader market have split into opposite succession regimes, each generating its own distinct form of leadership risk.

In the large-cap regime, the risk is entrenchment. When boards retain incumbents longer and require prior CEO credentials from replacements, they reduce the number of CEO-track developmental positions available. The COO, divisional president and group CEO roles that historically served as proving grounds stay occupied for longer or go to external candidates who already carry the credential. The funnel that produces future leaders narrows from above.

In the broader-market regime, the risk is readiness. When 84% of incoming CEOs are first-timers entering the role at a pace not seen in fifteen years, the question shifts from whether boards can find candidates to whether those candidates arrive adequately prepared.

The Pipeline Is Thinning from Below

The bifurcation alone would be a cyclical market observation. Two additional forces transform it into something structural.

Executive mobility has collapsed. LHH's 2026 survey of more than 2,500 companies found that the share of organisations experiencing high leadership turnover fell from 43% to 19% in a single year. Nearly 58% of late-career leaders reported no plans to leave within three years, up from just 11% the year before³. The senior executive layer that historically functioned as the developmental staging ground for CEO candidates has effectively frozen. Leaders who stay do not create openings for the cohort behind them.

Simultaneously, the next generation is stepping back from the pipeline. Deloitte's 2026 survey of more than 22,500 Gen Z and millennial workers across 44 countries found that only 6% named achieving a leadership position as their primary career goal. Nearly half of Gen Z respondents cited blocked advancement as a primary reason for leaving an employer⁴. These are stated preferences, not destiny — career ambitions shift substantially between twenty-five and forty — but they signal weakening pull at the bottom of the funnel at the same moment the top of the funnel is constricting.

These three dynamics — entrenchment above, a mobility freeze in the middle, and ambition withdrawal below — form a reinforcing cycle. Longer incumbency blocks pathways. Blocked pathways reduce mobility. Reduced mobility discourages ambition. Diminished ambition thins the pipeline. A thinner pipeline makes boards more reluctant to bet on untested candidates. That reluctance extends incumbent tenure further. The broader market breaks the cycle not by solving the capability problem but by absorbing its cost — installing first-time CEOs at historic rates and hoping that on-the-job learning substitutes for the developmental infrastructure that no longer exists.

Consider what this looks like inside a single large organisation. The CEO, now in year eight, remains effective and the board sees no reason to force a transition. The COO was hired externally three years ago with credentials designed to reassure investors. Two divisional presidents who would have been CEO candidates a decade ago have been in their roles for six years with no visible pathway upward. One leaves for a CEO appointment at a smaller company — where she joins the 84% of first-time leaders. The other scales back her ambition and optimises for tenure. Below them, three high-potential directors in their late thirties watch both departures, recalculate the odds, and invest their development energy elsewhere. The board's succession plan still lists internal candidates. The pipeline that was supposed to produce them has already hollowed out.

The Strongest Objection

The sceptical reading of this evidence deserves serious weight. The apparent bifurcation may be an artefact of comparing non-identical datasets with different methodologies, universes and possibly different time periods.

The Russell Reynolds data is labelled "H1 2026" but multiple citing sources note the figures may cover H1 2025. The exact measurement period is not clarified in any publicly accessible version of the index¹. If the experienced-CEO preference and the first-timer surge are sequential rather than simultaneous, the bifurcation thesis weakens substantially.

The datasets also overlap in ways that complicate the picture. The S&P 500 appears in both universes, and what Russell Reynolds counts as "prior public-company CEO experience" and what Spencer Stuart counts as "first-time enterprise CEO" are not necessarily complementary categories. Some appointments could be classified differently by each methodology without revealing any genuine divergence.

And the record experienced-CEO hiring rate of 23% still means 77% of incoming global CEOs lack prior CEO credentials. The claim that boards "demand proven capability" describes an intensifying minority preference, not a market-wide requirement.

These objections narrow the thesis without eliminating it. The bifurcation is more defensible as a description of diverging tendencies across market segments than as a clean partition into two discrete regimes. But even the narrower version carries a consequential finding: the forces that produce experienced-CEO demand at the top are causally connected to the forces that produce readiness gaps elsewhere, and most boards are preparing for only one of these dynamics while being exposed to both.

What the Tenure Research Complicates

Academic evidence on CEO tenure adds a dimension that cuts against simplistic conclusions in either direction. Research by Brochet, Limbach, Schmid and Scholz-Daneshgari, tracking S&P 1500 firms from 1992 to 2017, found a robust inverted-U relationship between tenure and firm value: performance rises for over a decade before declining after approximately fourteen years, with the inflection arriving earlier in dynamic industries characterised by higher R&D intensity or globalisation⁵.

Current average tenures — 8.5 years in the S&P 1500 and 9.0 years among large-cap global departures — sit well below that fourteen-year inflection point. By this measure, many departing CEOs may be leaving too early rather than staying too long. The entrenchment concern is not principally about value destruction by overstaying incumbents — not yet. It is about the pathway-blocking effects of long tenure and the credential concentration it produces, which operate long before the value curve turns.

For boards in the broader market, the finding is equally uncomfortable. Rapid CEO cycling — nearly 40% departing within five years — prevents leaders from reaching the performance zone that comes only with institutional depth. The system may be compounding readiness risk with tenure risk: appointing leaders who are less prepared and then replacing them before they have time to learn.

The Changed Question

The practical consequence of the bifurcation is not that boards need better succession plans. Most already have them. It is that the category itself needs to change.

Conventional practice treats CEO transition as an event: a departure triggers a search, a search produces a candidate, a candidate is appointed. The bifurcation reveals that the system producing capable candidates is degrading independently of any individual transition event. The pipeline problem is not a vacancy to be filled but a capability to be built — and rebuilt continuously, because the forces eroding it are structural rather than cyclical.

Boards that recognise this will measure executive mobility as a leading indicator of pipeline health, not as neutral HR data. They will design COO and group president roles explicitly as developmental positions with genuine profit-and-loss authority and board exposure, rather than as holding patterns for the next opening. They will quantify the gap between what their succession criteria demand and what their development systems actually produce — and be honest about the difference.

The harder question is whether any individual board can solve this alone. When the experienced-CEO supply is structurally finite and the developmental infrastructure is degrading across the ecosystem, the answer may require something the governance world has been slow to contemplate: cross-company investment in the leadership capability that every board needs but fewer organisations are producing.

The next time your board reviews its succession criteria, the relevant question is not only who meets them. It is whether your organisation — or any organisation operating under these dynamics — is still producing leaders who can.