Definition:Loss severity
📈 Loss severity measures the average dollar amount paid per claim within a given line of business, portfolio, or time period, and it stands alongside loss frequency as one of the two fundamental dimensions insurers use to characterize loss experience. While frequency tells a carrier how often losses occur, severity reveals how costly each individual event tends to be. Together, these metrics drive virtually every aspect of underwriting, pricing, and reinsurance strategy.
📐 Actuaries and underwriters calculate severity by dividing total incurred losses by the number of claims in a defined segment. The metric can be sliced by coverage type, policy period, geography, or any other dimension relevant to portfolio analysis. Severity distributions tend to be right-skewed — most claims cluster at lower values, but a thin tail of large losses can dramatically shift the average. To account for this, analysts often examine severity at various percentiles and model extreme outcomes using statistical distributions such as Pareto or lognormal curves, which feed into catastrophe models and excess-of-loss reinsurance structures.
🔍 Monitoring severity trends gives carriers early warning of shifts that could erode profitability before traditional loss ratio metrics fully reflect the change. Rising severity in auto insurance bodily injury claims, for example, may signal increasing medical cost inflation or shifts in litigation patterns — intelligence that allows underwriters to adjust rates proactively. In commercial lines, a sudden spike in average claim size within a professional liability book might prompt a review of policy limits and attachment points. Because severity directly influences reserve adequacy and capital allocation, it remains one of the most closely tracked indicators in any insurer's analytical toolkit.
Related concepts