Essay
The stuck employee does not exist. On who actually holds on.
“She has been here twenty years, you will never get her on board.” It is one of the most persistent assumptions in change work, and it has been measured. It does not hold. Where holding on to the existing does show up firmly is a more uncomfortable place.
When a change is planned, a list usually gets made. Who will come along, who needs convincing, and who you will probably not win over. That last group almost always contains the same people: the longest-serving employees.
The reasoning has two variants. Either they have gone stale after all those years. Or they sit in a golden cage: too many acquired benefits to still want to take a risk.
Both have been studied. Neither holds.
What has been measured
In 2013 Kunze, Boehm and Bruch looked at nearly three thousand employees for a link between years of service and resistance to change. They found none. Not weak, not in the expected direction: no link at all. It is the largest study on this, and there is no meta-analysis saying otherwise.
The broader stereotype that older workers are harder to move has been examined even more thoroughly. In 2012 Ng and Feldman brought together four hundred and eighteen studies, covering more than two hundred thousand employees, and tested six common age stereotypes. One survived: a lower willingness to retrain.
And even that one is not stubbornness. Someone with five years to go weighs a six-month training course differently from someone with twenty-five. That is arithmetic, not refusal.
The golden cage, finally, has no scientific basis at all as an explanation for resistance. The nearest studied concept, staying because leaving costs too much, predicts the opposite.
Where the pattern does come back
Here it gets uncomfortable. The same question, asked about the leadership rather than the staff, gives a different answer.
Across several studies, from Katz in 1982 to Hambrick and colleagues in 1993, longer tenure of directors and top teams is linked to holding on to the status quo, to a poorer fit between strategy and environment, and to teams becoming cut off from information coming from outside. Miller put it memorably in 1991: stale in the saddle.
Where you look for it
Is the long-serving employee holding the change back?
Where it sits
And the board that has been there a long time?
Repeated link
This is not meant as an accusation, and it is not one. Anyone who has been around a long time knows the house, and that is exactly what makes the starting point feel self-evident. You stop seeing the assumptions you made yourself. That happens to everyone who gets good at something.
But it does shift where you should look. The question “who is holding this back?” is usually asked downwards. The research says it belongs upwards as well.
What does predict it
If years of service are not it, what is? One mechanism holds up: how much someone personally stands to lose from the current situation. Accumulated switching costs, acquired rights, investments tied to a place or a role.
That is something other than tenure, even though the two often run together. Someone can be twenty years in and have little to lose from a redesign. Someone can be three years in and have just put everything on one card.
On top of that, “resistance” is usually the wrong word altogether. Reactions to change are not for or against. They are cognitive, emotional and behavioural at the same time, and often ambivalent: someone sees the point and feels bad about it at once. Piderit described this in 2000, and Oreg and Sverdlik confirmed it in 2026.
Read that as resistance and you are reading a deliberation as a position. And then you set out to convince where you should have listened.
What you do with it
Four things that follow, and that you can apply tomorrow.
Take tenure out of your assessment. The list of who will and will not come along is usually made on instinct, and that instinct is demonstrably miscalibrated. Ask instead, per role, what concretely disappears: which task, which piece of authority, which accumulated knowledge becomes worth less.
Ask the question about your own team too. How long has your board been in this composition, and when did something last come in from outside that questioned the starting point? That is the one place where the research does find the pattern.
Treat hesitation as information.Ambivalence means someone sees something you do not see yet. Anyone who says “I understand why, but this will clash with the autumn planning” is handing you a risk that was not in your plan. That is the cheapest quality control there is.
Design with the people who do the work. Not as a buy-in trick, but because they see where it chafes. That is also the first working principle of this approach, and this research is what underpins it: there is little resistance to route around, and a lot of knowledge to use.
The memory that plays along
One thing does predict how an announcement is received, and that is how the previous rounds ended. Not how much has already changed, but how well. That is a different body of research, and it is worked out here.
Together those two say the same thing. What happens in the room at an announcement is rarely about the people sitting in it. It is about what they have been through, and about what they stand to lose this time. Both of those are things you can design.