Zachry: Resistance to Change in Workers' Compensation: Bias and Incentives

Wednesday, June 17, 2026 | 0

Resistance to change in workers’ compensation is not a mystery. It is not structural inertia, regulatory complexity or organizational tradition.

Bill Zachry

It is the predictable result of two forces working in concert.

Human bias explains why people think the way they do. Financial incentives explain why they keep acting that way, even when they know better.

Bias, by itself, would eventually yield. New data, new leadership or a crisis costly enough to demand a response would force the system to adapt.

In practice, that is not what happens.

What we see

The signals are familiar to anyone who has worked in this system.

“We’ve always done it that way.”

“That’s our policy.”

“That’s how we handle this here.”

We see committees and delayed decisions presented as due diligence. We see cultures where people wait to hear the boss’s opinion before forming their own. We see selective data, avoided benchmarks and a quiet but determined resistance to comparison.

These are not random behaviors. They are predictable, and they are rewarded.

From the field: Safeway

When I arrived at Safeway as the new risk manager, there were 15,500 open files.

The settlement philosophy was to deny everything possible and then, when it was time to settle, to stipulate — provide lifetime future medical benefits, keep the file open and continue to fight everything along the way. It was the kind of approach that looks defensible from the outside but is ruinous from the inside.

I understood two things immediately.

First, the open inventory was generating massive collateral obligations. Every open file was a liability on the books, and the stipulation philosophy was manufacturing new ones faster than they were being resolved.

Second, and more important to me, this was not good for the workers.

An injured worker who remains tethered to the workers’ compensation system does not think of himself as a productive employee working toward recovery. He thinks of himself as an injured worker. That identity is not neutral; it is disabling, independent of the original injury. Every month a file stays open, that identity deepens. The system was keeping people sick by claiming to serve them.

On my first day, I told the entire staff that our claims philosophy was changing.

We were done fighting everything. We were going to bend over backward so far, we would have rug burns on our foreheads. Every claim that could be resolved would be fully settled through compromise and release. No more stipulations for manufacturing lifetime exposure.

But the philosophy was not just about settlement. It was about how we treated the people in those files.

I told the staff that we would treat every injured worker as if she were our own family member. That the medical care we arranged would be the best we could find, not the cheapest, not the most defensible, not whatever the fee schedule technically allowed. The best.

It is easy to manage a file. It is harder to look at a person — someone’s son, someone’s daughter — and ask whether you are actually doing right by them. I wanted my staff to ask that second question.

I also told the staff that if they settled claims and the caseload went down, nobody would lose their job. The reward for doing the right thing would not be a pink slip.

That changed everything.

When I told the staff about our new claims philosophy, one supervisor looked at me and said she had heard I was evil, but she hadn’t realized just how evil.

She meant it. She was not alone. The stipulation philosophy was not just an institutional habit; it was identity. For claims professionals who had built their careers around it, the new direction felt like an accusation, as if everything they had done before was wrong. Some of them were genuinely angry.

That reaction is worth understanding, not dismissing.

They were not bad people. They were people who had internalized a system’s values as their own. When the system said, “Fight everything and stipulate,” they fought everything and stipulated, and they became skilled at it and were evaluated on it. It became how they understood their professional worth.

The ones who came around did so when they understood two things: that no one would lose their job as caseloads declined, and that the new standard — treat every worker like your own child, find the best available care — gave them something more meaningful to be skilled at.

“Oh. We can do this.”

That moment was not enthusiasm for a management directive. It was people recognizing that the job they had always wanted to do was now the job they were being asked to do.

The supervisor who called me evil left the company a few months later. We did not fire her for what she said; that was never a consideration. But she was not ready for the changes that the new philosophy required, and she moved on.

Then, a few months after that, she called me.

She said she now understood what I had been talking about. She asked if she could come back.

That call told me more about the nature of resistance to change than anything I could have read in a management textbook.

She was not a bad claims professional. She was a capable person who had spent her career inside a system that rewarded a particular set of behaviors, and when those behaviors were no longer rewarded, her identity inside the organization became uncertain.

But she kept thinking about it. And when she could evaluate the philosophy from the outside, without the disruption of her own role being upended, she recognized it as correct.

Resistance to change is rarely about the change itself. It is about what the change implies about everything that came before it.

Give people time, remove the threat, show them results, and most of them will come around.

Some of them will even call you and ask to be part of it.

When I left Safeway, there were 5,000 open files. We had taken hundreds of millions of dollars off the collateral obligations.

And the workers whose claims were fully resolved had moved on — back to their lives, their work and their identities as something other than claimants.

What that experience reveals

What we experienced at Safeway was not unique. It was a clear example of how bias and incentives shape behavior across the workers’ compensation system.

The resistance we encountered was not irrational. It was predictable.

The biases that drive the behavior

These behaviors are not random. They are predictable.

Organizations default to what they already know. That is status quo bias.

They filter out inconvenient evidence. That is confirmation bias.

They grow confident in approaches that have not been challenged. That is overconfidence.

They avoid change in the name of prudence. That is zero-risk bias.

And in group settings, dissent is suppressed before it ever reaches the surface. That is herding.

Every organization has these tendencies.

That is not the problem. The problem is how the workers’ compensation system treats them.

The incentives that lock them in place

The workers’ compensation system does not merely tolerate these biases; it actively perpetuates them.

It organizes itself around them. It reinforces them. And most importantly, it pays for them.

In most parts of the system, individuals and organizations are not compensated for optimal outcomes. They are compensated for avoiding risk, following process, defending past decisions, managing appearances, containing budgets and hitting short-term metrics.

When those are the rewards, the safest path is not improvement. It is consistency — even when consistency produces inferior results.

What this looks like in practice

Changing a process introduces risk. If the outcome is worse — even temporarily — the decision-maker is exposed. If nothing changes, there is no blame. The rational choice is to do nothing.

Managers are measured on metrics they can influence by shaping how results are reported. The incentive is to highlight favorable data and suppress unfavorable comparisons.

Regulatory oversight and litigation risk reinforce the idea that mistakes must be avoided at all costs.

The system rewards being defensible, not being right.

In hierarchical organizations, agreement is safer than challenge. Ideas are filtered before they are spoken. Consensus becomes more important than correctness.

When closure rates and short-term cost-containment measures determine success, decisions are judged by results rather than the quality of the thinking behind them. This reinforces short-term thinking and resistance to change.

The uncomfortable truth

The system is not broken; it is working exactly as designed.

Not optimal outcomes. Not an efficient recovery. Not reduced human cost.

It is producing behavior consistent with its incentives.

Bias shapes decisions. Incentives reward those decisions. And the system hardens around both.

Over time, risk avoidance becomes a culture. Culture becomes policy. Policy becomes justification.

And change becomes something to be managed, not something to be pursued.

What needs to change

Better data will not fix this. Better technology will not fix this. Better ideas will not fix this.

What needs to change is what the system rewards.

Reward early and appropriate intervention, not just cost containment. Reward return-to-work outcomes, not just claim closure. Reward measured risk-taking, not just error avoidance. Reward transparency and benchmarking, not selective reporting. Reward the person who challenges the status quo, not the one who defends it.

Until that happens, the pattern will repeat: Good ideas identified, carefully studied, quietly resisted, not implemented.

The bottom line

People do what they are incentivized to do.

That is not a criticism. It is an observation.

The claims examiner who avoids a difficult conversation is not failing. She is responding rationally to a system that does not reward that conversation.

The manager who suppresses unfavorable benchmarks is not dishonest. He is protecting himself in a system that rewards the appearance of performance over its substance.

The organization that resists change is not incapable. It is behaving exactly as designed.

If we want different outcomes, we must change what the system rewards. Until then, the system will continue to produce exactly the results it is built to produce.

Not by accident, but by design.

Bill Zachry is a former board member of the California State Compensation Insurance Fund.