A zero that gets prettier when people get quieter is a filter, not a control.
I am Ryan L. Rinehart. I wrote Zero Harm Is a Hazard because I kept walking plants where the file looked world class and the job did not. This page is the short public record of that argument. It is not a sales page.
If you came here because you asked whether a zero harm goal improves workplace safety or hides incidents, the honest answer is: it can do both, and the hiding is more common than the brochure admits.
The goal is decent. Nobody wants a hurt person. The trouble starts when the goal becomes the scoreboard.
Zero harm, zero injuries, zero recordables, days since last incident on the board in the break room. Those are lagging counts. They tell you what already made it onto a form. They do not tell you whether the guard is still on the machine, whether the isolation held, or whether the last close call ever left the supervisor’s pocket.
The public critique is not new. Sidney Dekker has said for years that when you hold people accountable for a number near zero, you often get the number and lose the story. Fred Sherratt and Andrew Dainty looked at construction firms that wrapped themselves in zero language and firms that did not. The serious injury and fatality picture did not give zero a clean win. Andrew Hopkins has warned that a target of none can shift the work from learning to avoiding blame. OSHA’s own record on rate based incentives is blunt enough: when the prize or the bonus sits on a low injury rate, reporting gets quiet. Richard Fairfax’s 2012 memo and the later GAO work are still the federal paper trail for that.
I add one distinction the research rarely walks. There are two rooms.
The office room has the binder, the dashboard, and the days without a recordable. The floor room has the job as it actually runs on a Tuesday, with production behind and a crew that already knows which form will make them the problem. If you average those two rooms you get a lie with a decimal point. I do not average them.
Deepwater Horizon is the famous version of a clean lagging record sitting on top of a system that was not safe. You do not need a blowout to run the same pattern. You need a streak, a bonus, a customer who scores contractors on TRIR, and one person who would rather tape a cut than be the name that resets the sign.
So does zero harm improve safety? Sometimes the campaign funds real controls and the work gets better. Does it hide incidents? Yes, as soon as the number is treated as proof that the job is fine.
People stop reporting for ordinary reasons. Not because they hate safety.
They do not want to wreck the streak. They do not want the crew to lose the pizza, the jacket, or the quarterly bonus. They do not want to sit in a room that treats the report as a character defect. They are tired at the end of the shift and the process looks longer than the injury. They already watched a supervisor get questioned for “letting” a recordable happen while the schedule stayed sacred.
Call it peer pressure, embarrassment, fear, or inconvenience. The safety trade has a cruder name for the first aid that never hits the log: bloody pocket.
A zero injury target makes every report a failure of the target. If the company celebrates the absence of names, the person who puts a name on the page is the person who ended the celebration. That is not a mystery of human nature. That is what the system paid for.
The floor then learns a second lesson. Minor events vanish. Near misses stay in the break room. The office keeps printing a prettier file. The next event that should have been cheap information arrives as a serious one, because the cheap information never entered the system.
If your reporting went down after you hung the zero banner, do not congratulate the culture. Check whether the work got safer or whether the work got quieter.
Do not replace one slogan with a prettier slogan.
Use work you can watch this week.
Measure the presence of those things. Near miss quality. Time from a raised concern to a visible answer. How a supervisor talks when production is behind and the method has drifted. Whether a stop is treated as a save or as a delay.
You can still want nobody hurt. That wish does not need a banner. It needs a system that can stand a Tuesday.
I laid that out at length in Zero Harm Is a Hazard. The short version is the line on the cover problem: the audit was clean and the job was not. Start with the book if you want the full argument. Then walk your own last ten near misses and score the two rooms separately.
“If we do not say zero, we are saying some injuries are acceptable.”
No. You are saying the number on the sign is not the same thing as the control on the job. Accepting a slogan and accepting harm are not the same sentence.
“Our executives need a simple goal.”
Give them a simple goal that describes work. Three practices on the wall. One named asset. Two scores, office and floor. A rescore date. That is simpler than a number you cannot audit without scaring the reporter.
“Zero is a vision, not a target.”
If the vision is painted on the hard hat, counted on the board, and tied to a bonus, it is a target. Calling it a vision does not change the incentive.
I am not claiming zero language made every company less safe. I am not claiming lagging rates are useless. I am not promising that dropping the slogan drops your recordables. I am claiming that a goal of none, used as proof, trains people to protect the proof.
If you want the install of the model I use on one site, that lives on a different page. This page is the critique.
Ryan L. Rinehart, M.Ed.
EHS Concepts. The Ways of Safety.
thewaysofsafety.com
Zero Harm Is a Hazard on Amazon: https://www.amazon.com/ZERO-HARM-HAZARD-audit-clean/dp/B0HHXDZJ7Y