The Australian Financial Review recently examined one of the more awkward problems in governance: what happens when everyone except the chair thinks it is time for the chair to leave.
The stories are predictably uncomfortable. Directors meeting without the chair. Colleagues drawing the short straw and delivering the message. Chairs refusing to believe the feedback. Emergency conversations about who might take over. And, occasionally, the whole thing escaping into public view.
It makes for terrific copy - But our data suggests the real governance problem starts much earlier.
Across the BoardOutlook benchmark pool, most chairs are regarded reasonably well. Some 66.5% of respondents say the statement "The Chair provides effective leadership to the Board" is very accurate. Around 89% rate it at least moderately accurate.
So the destructive chair who has lost the confidence of the room and refuses to leave is real, but unusual.
The much more common problem is simpler. Chair transitions are difficult, often deeply personal, and rarely feel urgent until they are. But the reality of the data is that too many boards are failing to plan appropriately.
We ask directors and executives to identify the strongest aspects of chair leadership across 16 different attributes. The strongest are reassuring.
A good working relationship with the CEO is identified as a strength by 75.5% of respondents. Being well respected by the board comes in at 75.1%. Inviting participation from all directors is 66.6%.
Then the numbers begin to fall.
Having hard conversations when necessary is identified as a strength by 43.6%.
Providing feedback to directors falls to 20.6%.
And sitting at the very bottom is perhaps the most relevant capability of all: having a plan and timeline for chair succession.
Just 9.7% identify it as a strength.
At the same time, 17.5% nominate chair succession planning as an area requiring improvement, making it the single most commonly identified improvement area for chair leadership.
There is a fairly brutal conclusion hiding in those numbers. Whatever the reason, boards are least developed in the area that matters most when a chair transition becomes difficult: having a plan and timeline for succession.
The AFR article offers sensible advice about treating the removal of an underperforming chair as a governance and succession exercise rather than a personal confrontation. Quite right.
The problem is that this is much easier to say when there is already a succession process sitting underneath it.
If there is no agreed timetable, no conversation about tenure, no credible successor and no established expectation that the chair role will periodically turn over, the board is not really running a governance process. It is staging an intervention.
Boards understand succession. They plan for it, argue about it and revisit it, because they do it for the executive team as a matter of course. Across our benchmark pool, executive succession is visibly on the agenda: Boards are about two and a half times more likely to cite oversight of executive succession as a strength (24.6%) than the succession of their own chair (9.7%).
So the question is not whether boards know how to plan a leadership transition. It is a core part of the role, and they do it for the CEO and other key members of the leadership team. The question is why so little of that discipline gets applied to the person chairing the meeting.
Part of the answer is structural. There is no boss of the chair, no one whose job it obviously is to run the performance conversation. The chair often shapes board composition, board evaluations and the flow of information around the table, which is precisely why an orderly, pre-agreed succession plan should be the norm rather than the exception. A board that can map three years of CEO succession but cannot say who follows its own chair, or when, has left its single hardest transition to chance.
The AFR piece also identifies another reason chairs can become difficult to move on: they stop doing the uncomfortable parts of the job.
One director describes the "lazy chair", particularly one unwilling to confront director underperformance.
Our data gives that observation some weight.
Only 20.6% of respondents identify "provides feedback to Directors" as a strength of their chair. It is also nominated as an improvement area by 17.7%.
This matters because chairing is not just meeting facilitation. The chair has to build the board as well as run it.
That means dealing with the director who no longer contributes enough. The director who dominates. The director whose expertise is becoming less relevant. The director who does not prepare properly. The director who has stayed too long.
None of those conversations are particularly enjoyable. Avoid enough of them for long enough and the board slowly accumulates problems. Eventually somebody has to deal with them, except by then the question may no longer be whether a director should leave. It may be whether the chair should.
There is something telling about the stories in the AFR piece.
Directors gathering after dinner without the chair. Someone volunteering to deliver the bad news. The board hastily working out who might take over. That is not really succession. That is succession planning having failed.
The best chair transitions should be considerably more boring.
There should already be an agreed view about tenure.
The chair should receive regular, credible feedback.
Director performance should be discussed before it becomes a crisis.
The board should understand what capabilities it will need next.
Potential successors should not be a subject first considered five minutes after somebody has finally told the incumbent to leave.
None of this eliminates the possibility of a chair digging in. Human beings remain inconveniently human.
But it changes the nature of the conversation. Instead of: "We have decided you need to go." It becomes: "We have reached the point in the succession process that we have been discussing for the past two years."
That is a very different conversation.
The point is not to make difficult chair transitions disappear. It is to make sure they do not arrive as a surprise.