Solvency II internal model
An insurer's own model, approved by the supervisor, used to calculate its solvency capital instead of the standard formula.
Directive 2009/138/EC (Articles 112 to 127) lets an insurer that finds the standard formula a poor fit for its profile compute its SCR with a model of its own, full or partial, once the supervisor agrees; the supervisor has six months to decide. Calibration remains the Article 101 standard, a 99.5% VaR over one year. What changes are the assumptions, the dependence between risks and the catastrophe tail. Articles 120 to 125 set out what the supervisor checks, and the first carries the most weight in practice: the use test. A model that drives the capital number without driving underwriting and reinsurance decisions does not pass. Capital is often the motive: a well-calibrated model on a better-than-average book produces a lower charge than the standard formula. That is also why approval is strict and validation annual. In cyber, the standard formula has no notion of the risk as such. It runs through the non-life premium and reserve risk of whichever line carries it, and through operational risk, with no catastrophe module. The internal model is the only place where an insurer puts its cyber accumulation into capital, and it then bears alone the burden of calibrating a risk with almost no history.
Textbook case: a large cyber reinsurer develops a partial internal model for its underwriting risk, on the view that the standard formula underestimates correlation among insureds that rely on the same providers.
internal model, partial internal model, Solvency II internal model
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.