Theoretically, if the company has many accidents, it’s a function of time before they have a bad, or even catastrophic, one. It may just be bad luck to have one injury, and it’s a bad one.
Personal protective equipment is designed to reduce the severity of injuries while safety training tends to reduce the frequency. Therefore, safety training, job ergonomics, and awareness reduce injuries, and workers’ compensation costs, most.
The experience rating formula reflects these philosophies. The rating system compares expected losses to actual losses and sets a modification as a variance from that norm.
The “exposure unit” for workers’ compensation is payroll in one hundred dollar increments. Historical losses for an industry are compared to historical payrolls to determine a rate of loss per hundred dollars of payroll. A severity discount is applied to the expected losses. This rate is not your premium rate, just the rate of expected losses. The difference in the two rates includes the cost of claims service, loss control, administrative costs, sales commissions and taxes.
Actual losses below a certain severity threshold, and that varies state to state, are added together. In addition, severe claims are individually discounted on a scale from the threshold rate to a state maximum. These claims are added to the undiscounted frequency claims value. The resulting number is compared to the theoretical average expected losses as actual over expected. A three year historical period is used.
The resulting ratio is the experience rating. Poor experience will result in a number greater than one. Good experience, below one. For example, a company has an experience modification of 1.15. That company is paying fifteen percent more for workers’ compensation than his average competitor and forty three percent more than a competitor with a .8 modification.
That’s additional cost of labor. That’s a more satisfied worker who isn’t at risk of injury. That’s a competitive edge that safety provides by reducing the frequency of injuries first.
