Question ID: 1537
Regulation Reference: (EU) No 2015/35 - supplementing Dir 2009/138/EC - taking up & pursuit of the business of Insurance and Reinsurance (SII)
Article: 121
Status: Final
Date of submission: 14 May 2018
Question
- Is it currently a requirement under Solvency ii legislation that windstorm clustering must be taken into account when estimating exposure to aggregate European wind risk through European windstorm catastrophe models?
- If so, from what date did it become a requirement? And where is this published?
- If so, what was the rationale behind the requirement that windstorm clustering must be taken into account?
- Was the work of Professor David Stephenson (University of Exeter) into catastrophe modelling of storm clustering an influencing factor on the inclusion of this requirement within the regulatory framework?
- If not, are there plans to include it as a requirement? And when will this occur?
EIOPA answer
Article 121 of the Commission Delegated Regulation 2015/35 defines how capital requirements for windstorm risk should be calculated. Two scenarios are defined: the first with two moderate events; the second with one large and plus a second smaller event.
There is no explicit requirement that windstorm clustering should be taken into account, although if it is a relevant and material risk, insurers and reinsurers would be expected to reflect it in their ORSA.
These two scenarios were defined by the CAT Task Force, composed of representatives of different background including modellers, reinsurers, brokers and supervisors. The CAT Task Force reflected the information, models and academic literature available at that time (2009-2010). However, to the best of our knowledge, there was no reference to the work of Professor David Stephenson in particular.
The document you are referring to is a consultation paper that was published in the context of the review of the capital requirements for (re)insurers. After considering stakeholders’ feedback, it was decided not to further investigate windstorm clustering for the standard formula at this stage.