Solvency II’s one-year time horizon: A case for including the full variance

Traditionally, non-life reserving risk considers risk over the remaining lifetime of liabilities, which in early models was quantified via approaches that focused on the standard deviation of the outstanding reserves, including uncertainty for both parameter risk and process risk. Under Solvency II, non-life reserving risk takes on a different meaning, based on the change in the estimated ultimate loss over a one-year time horizon, which accounts for the payments during the one-year time horizon and the consequences for future payments after the one-year time horizon. A number of models have provided insurers well-thought-out and documented approaches for determining reserve variability and estimating unpaid claims on an ultimate time horizon and a one-year time horizon, respectively.