Capping (large loss capping)
Actuarial technique that caps large losses at a threshold to isolate their impact when analyzing attritional loss experience.
Pickands, 1975, Annals of Statistics: above a high enough threshold, excesses approximately follow a generalized Pareto distribution. That result is what makes capping workable. Each claim is limited to a threshold (the cap) during loss experience analysis, and the portion above the cap is modeled separately, with extreme value methods or a dedicated tail distribution. Attritional experience, frequent and moderate, is thereby split from large losses, rare and expensive: the two follow different statistical laws and are priced separately. The cap moves the calculated severity and pure premium directly, so a loss metric is always reported with the cap used.
Textbook case: on a liability book, a 500,000 EUR cap is applied. Three claims above the cap (900K, 1.2M, 2.1M EUR) are handled separately. Attritional severity on capped losses is 18,000 EUR vs 47,000 EUR uncapped.
Follow the research
One letter a week. First, the week's underwriting note, when one is published: a dated, argued decision, signed by Paul Bernardski. Then what the observatory published on cyber risk, political risk and insurance. No account, no tracking.
No spam, one-click unsubscribe, no data ever sold.