General liability classification shouldn’t be a guessing gameBy Rick Stoll | March 25, 2016
Other than historical loss experience, there are three fields on an insurance application that lay the groundwork for a general liability quote: address (territory), exposure (sales, payroll, square footage, etc.), and GL class. Each field has its own effect on the underwriting process, but the GL class is arguably the most subjective. Verisk Insurance Solutions defines about 1,000 class codes to determine an appropriate loss cost. The classes and associated loss costs are continually refined and recalculated to make sure insurers match price to risk fairly. Price-to-risk accuracy is essential to competitiveness, customer retention, and profitability.
In some cases, agents won’t have enough information to classify a business. As a substitute, they’ll provide the SIC number or NAICS code, make a best guess, or use an industry catch-all class.
None of those options is a good substitute for an accurate classification. It’s imperative that underwriters are confident they’ve chosen the appropriate class for each quote because the consequences can be significant. For example, in Jersey City, New Jersey, the riskiest restaurant class commands more than five times the expected claim cost and associated premium as the least risky class. That type of price variance can significantly and negatively affect profitability through adverse selection and lost revenue opportunities.
Understanding the effect of proper classifications doesn’t make an underwriter’s job any easier. They’re constantly asked to leverage data and analytics to generate greater revenue and profitability—the classic “more with less” approach. The key is to use the right data and analytics to streamline the class selection. An industry selection from a well-respected, objective third party provides a tremendously valuable data point. Having additional details on such factors as business description, liquor license, or OSHA violations can increase an underwriter’s ability to differentiate confidently between two seemingly similar classes with materially different loss costs.
Tagged with: , , ,