Why Middle Managers Resist Change

Change

Change Everything. Just Not Us.

Oct 28, 2026

8 min read

The sergeants who ran the city

By the early 1970s, corruption in Hong Kong was not a scandal. It was infrastructure.

Police stations were organized to make money from street hawkers, licenses and gambling, and the payments moved through the ranks with the regularity of a payroll. The people who ran it were mostly not the senior officers at the top. They were the station sergeants in the middle, who knew every shop, every arrangement and every price. By one investigator's account, they grew enormously rich.

The police had an anti-corruption branch of its own. It achieved very little, which surprised no one. It was asking the system to investigate itself.

That changed in 1973, when Peter Godber, a senior British officer under investigation, slipped out of the territory. Public anger forced the government's hand. In February 1974 it created the Independent Commission Against Corruption, which answered to the Governor and to no one in the police.

Then the middle fought back. In 1977, as investigators closed in on the syndicates, officers protested, marched on the commission's headquarters and pushed the force toward something close to mutiny. On 5 November, Governor Murray MacLehose announced a partial amnesty. Offenses committed before 1 January 1977 would no longer be pursued, apart from a few serious exceptions.

It looked like surrender. It turned out to be the move that made the change possible. The past was forgiven, the future was not, and the line between the two was drawn in public. Over the following years, one of the most corrupt police forces in the world was gradually rebuilt, and Hong Kong became known as one of the cleanest places in Asia to do business.

Two other countries, two other routes

Georgia chose a harsher path. After the Rose Revolution, its traffic police were among the most despised institutions in the country, known less for enforcing the law than for collecting bribes at the roadside. In 2004 the new government did what most reformers only joke about. It dismissed the entire traffic police force and started again, with new recruits, new training and salaries people could actually live on.

For a few weeks there was no traffic police at all, and many observers felt the roads were no worse for it. Undercover checks followed, so the new force knew it was being watched. Even the government's fiercest critics ended up praising the result.

Singapore started at the top. The Prevention of Corruption Act of 1960 gave the Corrupt Practices Investigation Bureau unusual reach, including letting courts treat a lifestyle an official's income could not explain as evidence of bribery. But the real signal came from who the bureau was willing to pursue. In 1986 it investigated a serving cabinet minister and close associate of Lee Kuan Yew. After that, nobody lower down could pretend the rules were written only for them.

Three countries. Three very different methods. The pattern underneath is the same.

None of them succeeded by changing the rules alone. Each changed the layer that decided whether the rules meant anything.

The company that kept changing nothing

I was once called into a company to answer a deceptively simple question: why was it failing?

Leadership wanted change. That was never in doubt. They had tried more than once. New structures were announced, new processes rolled out, new values printed and framed on the walls. Each attempt began with energy and ended in the same quiet place, with an organization that looked slightly different and worked exactly the same.

What I found was not complicated. The resistance was not at the top, and it was not on the floor. It sat in the middle.

Middle management was perfectly willing to accept change, as long as it was the kind that did not touch them. A new logo. A renamed department. Another round of training for the people below. Anything that reached their own authority, their own routines or their own corner of the org chart was slowed, softened or quietly absorbed.

They were not villains. They were doing what Hong Kong's sergeants did, protecting a system that had served them well. And every previous attempt had repeated the mistake of the old police anti-corruption branch. It had asked the middle to lead a change that was aimed at the middle.

Change everything, the message had been. Just not us.

What the research says about the middle

This is not one company's problem. Prosci, which has studied organizational change for decades, regularly asks practitioners which group resists change the most. Mid-level managers come out on top, ahead of front-line employees, and in its 2023 research 43 percent of participants named them as the most resistant group.

The reasons are human, not ideological. Managers often lack a clear picture of what the change means for them. They are rarely told what is in it for them. Many doubt they have the skills to lead their own people through it. And where senior leaders do not visibly back the change, managers feel exposed, so they wait to see whether it will last.

Prosci's ADKAR model makes the trap easier to see. Change has to pass through five stages in every individual: awareness, desire, knowledge, ability and reinforcement. Most failed programs pour their effort into awareness and knowledge, the announcements and the training. Cosmetic change survives precisely because it never asks for desire. It asks nothing a manager would actually have to want.

John Kotter reached a similar conclusion from another direction. In his study of why transformations fail, he warned against leaving powerful obstacles in place and against declaring victory before change sinks into the culture. The middle layer is often both at once: the obstacle nobody wants to name, and the culture the change was supposed to reach.

The irony is that the same research points to managers as the strongest allies change can have. They sit closest to the people who must change, and Prosci finds that an employee's direct supervisor is the most effective source of reinforcement during change.

The middle is where change usually dies. It is also the only place it can live.

What actually moves the middle

These countries shared no culture, no political system and no playbook. What they shared were a handful of moves that work just as well inside a company.

Take the change out of the hands of the people it is meant to change. Hong Kong's breakthrough was a body that answered to the top, not to the police. In a company, that means a sponsor with real authority and what Kotter calls a guiding coalition: one that includes people from the middle without being owned by them.

Start at the top, and make it visible. Singapore's message became believable the day it pursued a minister. Managers watch what senior leaders change about themselves before they change anything about their own teams. Prosci consistently finds active, visible sponsorship to be the single biggest contributor to successful change.

Make it personal for the middle. Before asking managers to lead a change, help them through it as individuals. Tell them plainly what they will lose, what they will gain and what will be expected of them. Desire cannot be announced. It has to be earned, one manager at a time.

Draw a clear line between the past and the future. Hong Kong's amnesty was controversial, and it worked. People defend the past hardest when they fear being punished for it. Give the middle an honorable way across, then be immovable about what happens on the other side.

Remove what still blocks the road. Georgia is the extreme case, and very few companies need anything like it. But most have a few roles, reporting lines or individuals that will keep absorbing change however well it is explained. Leaving them in place quietly tells everyone else the change is optional.

Anchor it. Georgia kept testing its new police long after the launch. Change becomes culture only when the way people are hired, promoted and rewarded begins to reflect it. Until then, it is just an announcement.

This is where most of my work happens. Not in designing the change, which is usually the easy part, but in the layer that decides whether it ever arrives.

A question before your next change program

Hong Kong did not become clean because it wrote better laws. It already had laws. It became clean when the sergeants understood that the old arrangement was over, and that there was a way for them to survive its ending.

Every organization that has failed to change more than once has its own version of those sergeants. Usually they are not corrupt. They are capable, loyal people protecting a world that made sense to them.

So before you launch the next initiative, ask yourself honestly:

Who in your organization is quietly approving change, as long as it happens to someone else?

And what would it take for them to want it too?

It's time to evolve.