No one would argue that in the past few years almost every business has struggled through tough economic times. The economy has hit workers' compensation insurers particularly hard. Given that workers' compensation premiums are based on employee payroll, the insurance industry would expect to see a similar decline in payroll as that of its premiums. However, according to the Bureau of Economic Analysis, wage and salary disbursements had a net increase of 3.5% over the past three years.

While the rate decreases that workers' compensation insurers have made over the years can explain a portion of the discrepancy, they do not explain the majority. The gap between payroll disbursements and written premium is widening, leaving almost one half a trillion dollars of payroll unexplained to insurers. This widening gap is making insurers question if they are underwriting the actual exposure, and if that exposure is classified correctly.

So, what's an insurer to do? While conducting field audits for all new business during underwriting and for each subsequent renewal is a foolproof way to remedy the problem, this approach is both cost- and time-prohibitive. However, there is a way for insurers to separate the vital policies that require closer inspection-predictive analytics.

By employing vast external databases and sophisticated statistical analysis, some insurers are now using predictive models to not only identify which policies merit audits, they are also using predictive models to determine the most cost-effective audit; whether it be a physical audit, a telephonic audit, a verified telephonic audit, or a voluntary audit.

Under the Table

One area of particular focus for predictive models is that of the underground economy. To manage costs, companies are relying more heavily on sub-contractors to maintain a lean employee workforce. The magnitude of the problem is speculative; however, it has become so acute that many states have enacted task forces to address the mounting problem. Some states estimate as much as 10% of payroll as unreported. Workers' compensation is a $40 billion industry, so this estimate would translate into $4 billion in premium leakage. With predictive models insurers are better able to identify those policies likely to have uninsured sub-contractors, offering the much-needed true risk insight to reduce this premium leakage.

This downturn in the economy is causing companies to adapt their business focus. Companies that classically operated in sectors that were hit particularly hard by the economic downturn, such as new home construction, are quite often modifying their business operations to focus on other areas, such as plumbing and HVAC.

With these business changes comes a changing risk profile of an insurer's renewal book of business.

Perhaps of biggest concern to insurers are those companies that find creative ways to report the classification of employee payroll to reduce premiums. A recent analysis of a large, multi-carrier database of workers' compensation field audits showed that 10% of policies had more than 30% of the exposures misclassified. Not surprisingly, the misclassifications were biased toward classes of risk whose rates were substantially lower than that of the company's true risk class. Whether the misclassification of employees is accidental or fraudulent, the lessons for workers' compensation insurers can be all-too-painful to their financial results.

Take, for example, a workers' compensation insurer that underwrote a Nevada manufacturer under the governing class code of a sporting goods manufacturer. After an explosion ripped through the plant, killing one and severely injuring others, the insurer began to ask questions. The resulting investigation quickly highlighted that the company was not a sporting goods manufacturer, but rather a manufacturer of miniature rockets and missiles; a class of risk whose workers' compensation rate is nearly ten times that of a sporting goods manufacturer. To further exacerbate the problem, the insurer had specific treaty reinsurance exclusions for rocket and missile manufacturers. In this circumstance, the insurer unknowingly wrote business outside of their underwriting appetite and reinsurance umbrella protection.

Many states have regulatory restrictions that prevent insurers from adding classes of risk to a workers' compensation policy until the subsequent policy renewal. Since these missing classes are typically discovered after a policy expires, insurers are often burdened with the original risk classification until the third year the policy is on their book of business. As a result, more and more insurers are employing predictive models much earlier; when the policy is first underwritten, to ensure that the policy is classified correctly from the start.

With predictive analytics insurers have a fighting chance to better understand the right risk in order to get the rate right-the perfect formula for a profitable and sustainable workers' compensation book of business.

This article can also be found at

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access