Outgrace
OutgraceThe Leadership Magazine, home
LeadershipData brief

Leader tenure and change: what the evidence does not yet show

Five sources searched; none links how long a leader stays in post to whether a change programme succeeds. What boards should do instead.

Boards ask a predictable question when a transformation stalls: is the sponsor still the right person to lead it, and how long has she or he held the job? It is a fair question. The research we checked cannot yet answer it. We searched five sources, among them the World Bank, the National Bureau of Economic Research, Harvard Business School, Stanford's Graduate School of Business and Knowledge at Wharton, for any study linking a leader's time in role to whether a change programme succeeds. None reports one. Nothing in this piece shows that tenure matters, and nothing shows that it does not.

Alumni claimed in HBS learning networks, self-reported; no link to change outcomes
86,000
Start of Helena Foulkes's at-large NBER board seat; biographical date, not outcome data
April 2020

What the data shows

The most substantive material we found is an episode of Knowledge at Wharton's podcast, This Week in Business. Its guest, Henning Piezunka, Wharton Associate Professor of Management, discusses CEO succession and the risks of centralised leadership. The episode is interview commentary, not measured results. It gives no tenure durations and no outcome figures, so it cannot tell a reader how long leaders last or what happens to their programmes.

Within the episode, the points concern succession more than tenure. Piezunka, according to the summary, argues that companies need organisational resilience beyond their most important executives. The same discussion examines how hard it is to groom future chief executives, and it names the power of corporate boards as a factor in succession. These are reasonable claims to test. The episode does not test them.

The discussion cites leadership transitions at Apple, Disney and other major organisations as examples. It frames the leadership turmoil at NVIDIA as a question of whether a company built around one leader can survive without that leader. The episode poses the question. It does not answer it with evidence, and the examples are not presented as a sample of change programmes with measured results.

Harvard Business School describes programmes for experienced executives who want to lead at a higher level. Its site gives no outcome evidence for them. It also claims 86,000 alumni in its learning networks. The school reports that figure itself, and the site does not connect it to change outcomes. A large alumni network says something about reach. It says nothing about whether a sponsor's time in role affects a transformation.

Stanford's Graduate School of Business states a mission of developing leaders who change the world. The mission statement cites no tenure research and no change research. Wharton Executive Education and Wharton's Center for Leadership and Change Management jointly offer leadership tools that they describe as fast and effective. The sources do not describe what those tools contain, so we cannot say whether they touch on tenure at all.

The National Bureau of Economic Research's news item concerns Helena Foulkes, an at-large NBER board member since April 2020. It reports that she was elected Governor of Rhode Island on November 3, 2026, with more than 57 percent of the votes. It describes her as former chief executive of the Hudson's Bay Company and records nearly 25 years at CVS Health, including roles as President of CVS Pharmacy and Executive Vice President of CVS Health. It is biography. It contains no outcome data on any tenure.

Figures in the sources, and what they measure

None of these figures measures leader tenure or change outcomes. They appear because the sources state them, not because they answer the question.

Alumni claimed in Harvard Business School learning networks; self-reported, not linked to
86,000
Votes received by Helena Foulkes in the Rhode Island governor election; not tenure data
more than 57 percent
Helena Foulkes's career at CVS Health, as recorded by NBER; biographical, no outcome data
nearly 25 years

Source: hbs.edu; National Bureau of Economic Research

What it means for a leader

Three things follow. First, a board paper that says long-serving leaders improve or damage transformations would claim more than these sources support. Second, the question beneath the tenure debate still matters: can the programme survive a change of sponsor? The Wharton discussion points that way, and the question can be tested without a published study. Third, the leadership cases in these sources are examples of change at the top, not tests of transformation outcomes. A programme's result at Apple, Disney, NVIDIA, the Hudson's Bay Company or CVS Health cannot be read from these pages.

In our judgement, a sensible test is this. List the five people without whom the programme would stall. For each, write down who could take over within three months and what that person would need to know. If the list has one name against most of the roles, the risk sits with one person, whatever that person's years in post. Put the list in front of the board before the milestones, not after the sponsor's first absence.

Write the succession question into the transformation plan before the programme starts. Many boards treat succession as a separate governance exercise, which may leave it silent about the programme itself. The Wharton discussion names the power of boards over succession as a factor. That is a reason to bring the board in early, so that the sponsor's handover appears in the same document as the milestones and the budget.

Resist the rule of thumb. A fixed number of years in role, or a belief that a new sponsor always resets momentum, is easy to state and hard to defend. None of the five sources supports either claim. A rule with no evidence behind it is also difficult to explain to a board that later asks what the company's own records show. The better course is to look at the company's programme history and see whether any pattern exists there.

Limits of the data

The limits are severe and should be stated plainly. The five sources were searched for tenure and outcome data, and none contains it. The Wharton episode is commentary from one academic, not a measured sample. Harvard's alumni figure is self-reported and unconnected to outcomes. The NBER item is biographical, and the Stanford statement cites no research. No source disagrees with another on tenure and change success, because none addresses it; silence is not agreement. For a board that needs evidence, the next step is an analysis of the company's own programme records, with tenure and outcome defined in advance and reviewed by someone trained in research. If a leadership change carries legal, financial or employment consequences, take advice from counsel or an employment specialist before acting.

Newsletter

The Briefing

One email a week: the leaders we profiled, the ideas worth stealing, and nothing else.

Weekly, free, one click to leave. Read our privacy policy.