Loss-sensitive rating plans and solvency capital under solvency II
Applied Mathematical Sciences · 2026 · 20 (2) · 61-68
DOI: 10.12988/ams.2026.919319 ↗Kurz gefasst
Fachartikel in der Zeitschrift Applied Mathematical Sciences, 2026. Verfasst von Antonella Campana und 1 weiteren Autor:innen.
Abstract (Originalsprache)
Loss-Sensitive Rating Plans (LSRPs) introduce explicit risk sharing between insurer and insured by linking premiums to realized losses. Recent actuarial research has shown that retrospective rating plans, large deductible plans and stop-loss arrangements can be represented within a unified framework based on layered loss functions and stop-loss transforms. A key implication of this framework is that pricing rules based solely on expected values fail to capture the variability and tail risk borne by the insurer. This paper embeds the loss-sensitive rating framework within the Solvency II regime, where capital requirements are calibrated to tail risk through a Value-at-Risk measure at the 99.5% confidence level. We show that loss-sensitive contracts primarily affect solvency capital by reshaping the upper tail of the insurer’s loss distribution rather than by reducing expected losses. We numerically assess how different contract designs with similar expected outcomes can lead to substantially different capital requirements. The results highlight the need to integrate pricing, tail-risk measures and capital considerations when evaluating loss-sensitive rating plans.
Zitieren
APA
Campana, A., & Sardellitti, S. (2026). Loss-sensitive rating plans and solvency capital under solvency II. Applied Mathematical Sciences, 20(2), 61-68. https://doi.org/10.12988/ams.2026.919319
Vancouver
Campana A, Sardellitti S. Loss-sensitive rating plans and solvency capital under solvency II. Applied Mathematical Sciences. 2026;20(2):61-68. doi:10.12988/ams.2026.919319
BibTeX
@article{Campana2026,
author = {Campana, Antonella and Sardellitti, Stefania},
title = {Loss-sensitive rating plans and solvency capital under solvency II},
journal = {Applied Mathematical Sciences},
year = {2026},
volume = {20},
number = {2},
pages = {61--68},
doi = {10.12988/ams.2026.919319}
}
EUVIMED