The Safety Incentive That Made Near-Misses Disappear
Prairie Metalworks is a 260-employee fabricator. Eighteen months ago the plant introduced a quarterly safety bonus: if a department records zero recordable injuries for the quarter, every employee in it receives $250. Over the last four quarters, recordable injuries fell from 14 to 4 - but the trend is not reassuring. Near-miss reports dropped from about 40 a quarter to 6. Two employees have privately told the safety coordinator that a coworker cut his hand badly in March and 'didn't want to make a thing of it,' and that crews now sort out minor injuries among themselves. One department has posted a whiteboard tracking 'days without an incident.' The plant manager considers the program a success and wants to raise the bonus to $400.
1. Is the drop in recordable injuries credible? What is actually happening, and why is it dangerous?
Read the worked answer
The leading and lagging indicators are moving in opposite directions, and that is the classic signature of underreporting.
- In a functioning safety system, near-miss reports rise as reporting culture improves, and injuries fall. Here near-misses fell 85% while recordables fell 71%. Hazards don't disappear that fast - the reporting is what disappeared.
- The incentive design is the cause. A $250 bonus per person conditioned on zero recordables means an injured coworker costs the whole crew money. That is a direct financial pressure on peers to conceal injuries and on supervisors to classify them as first aid rather than recordable. The whiteboard 'days without an incident' reinforces it - it makes a report a visible setback for everyone.
- Why it's more dangerous than the old rate: the plant has lost its early-warning system. The March hand injury is exactly the case that should have generated a report, an investigation, and a machine-guard fix. Instead there is a hidden injury, a possible recordkeeping violation, and an employee who may now have an untreated or worsening injury and a workers' compensation problem.
- It may also be a recordkeeping exposure: if injuries occurred and were not recorded, that is a violation independent of the incentive.
2. What should HR and safety recommend doing with the program, and what should replace it?
Read the worked answer
Do not raise the bonus to $400. The incentive is the problem, not the dose.
- Immediate actions: stop disqualifying departments for reported injuries. Interview the March injury privately and off the record, confirm whether the employee received treatment, and get him care. Pull the OSHA 300 log and reconcile it against first-aid logs, clinic visits, and workers' comp claims for the last four quarters.
- Replace the outcome-based bonus with a leading-indicator program. Reward the behaviors that actually reduce injuries: near-miss reports submitted and closed, hazard corrections completed on time, toolbox talk participation, and inspection findings resolved. A department that reports 20 near-misses and fixes them is safer than one that reports zero.
- Remove the 'days without an incident' board. It converts a report into a public failure. Replace it with a board of hazards found and fixed.
- Say it plainly to the plant manager: the incentive is buying lower numbers and selling higher risk. A single serious injury that was concealed to protect a $250 bonus is a claim, a citation, and a morale collapse - and the plant manager's own record is on it.
- Expect near-miss reports to jump when this changes. Brief leadership in advance so the increase is read as the program working.
3. The plant manager says the incentive works because injuries went from 14 to 4, and asks how else you would reduce injuries without paying for it. How do you respond?
Read the worked answer
Take the question seriously - he has a budget argument, and it deserves a budget answer.
- Reframe the metric. He is measuring the number he can influence by suppressing reports. Offer to be judged on a different set: recordables plus near-miss reports, near-miss closure time, and repeat findings from inspections. If injuries fall while near-misses are being reported and fixed, that is real. If they fall while reporting falls, it isn't.
- Cost the alternative. The $250 program across 260 employees is roughly $65,000 a year if all departments hit it; at $400 it is over $100,000. A near-miss reporting and hazard-correction program costs a fraction of that, and it targets the causes of the injuries rather than their paperwork.
- Show the hidden cost side. One concealed hand injury can produce a workers' comp claim, an OSHA recordkeeping citation, and - if it becomes known that employees were pressured not to report - a retaliation exposure. That exceeds the annual bonus budget.
- Reassure him on the thing he actually cares about: this is not a proposal to spend more, it is a proposal to spend the same money on leading indicators. A plant that finds hazards early has fewer shutdowns, less unplanned overtime, and fewer lost-time injuries on his record.