Improving natcat modelling - where can we find new opportunities?
Article information and share options
Global catastrophe experts gathered to explore a number of significant opportunities in natcat modelling for the insurance industry at the inaugural Catastrophe Knowledge Exchange, held at the Swiss Re Centre for Global Dialogue.
Dickie Whitaker, the Chief Excutive of Oasis Loss Modelling Framework, discussed using open source platforms as a potential way for modellers to leverage knowledge and expertise within the natcat modelling community.
During the discussions about these platforms, much emphasis was placed on data standards and the need to avoid having too many initiatives focused on the same outcome. While the need to collaborate was mentioned, concerns were raised about the risk of engaging in groupthink about how models should work. Nevertheless, more openness is needed, particularly on how models are created. Common data sets would also be appreciated.
The link between natural catastrophes and business interruption (BI) was also explored at the event. Otto Kocsis, Global Head of Business Resilience Technical Center at Zurich Insurance, noted in his session, "CBI/BI - where to go next?" that the business interruption share of property claims has doubled in the last 10 years due to factors such as globalisation, different production patterns, global value chains, fragmented value chains and lean value chains.
In his role, he seeks to understand and model this exposure. While some industries are more exposed to BI than others, it is important to understand which industries are located at sites where natural catastrophes could occur and how redundant the exposure patterns are.
Insuring the obvious
In Martin Bertogg's discussions "How will insured risks change - what will we insure in the future?", the Head of Property Treaty Globals EMEA, at Swiss Re, looked less on new developments that could potentially affect the insurance sector such as cyber risk and the sharing economy.
Instead, the discussions were centered on how to close the existing protection gap in insurance – i.e. insuring the obvious. Models are already used very successfully in optimising the way the acquired risk is being handled, but very little of this knowledge is used for optimising and innovating the products sold. Are insurers too conservative? Property policies, for example, have not changed for decades in regards to terms and cover. How can insurers grow the business? A rise in the number of catastrophic events would stimulate demand for insurance. Uptake is also likely to increase if governments take a more active role in educating consumers and promoting the benefits of insurance.
The topic of how to make smart use of available data was also presented Martin Spoerri, Head Info Foundation and Smart Analytics at Swiss Re, and Beat Aeberhardt, Heat of NatCat Tools, Swiss Re. Swiss Re is looking at how it can best use the exposure data it has been collecting from its natcat models to build a new data set where every single property risk is identifiable.
There are some challenges that need to be overcome, such as the data quality of the exposure information. Moreover, it is not yet clear if this data be shared from a legal perspective. It was also mentioned that increased regulation and Solvency II will have a significant impact on data quality in the future.
Future improvements in location intelligence
In "Future improvements in location intelligence" breakout session, by Mathias Künzler, Head Natural Hazards, Mobiliar, and Peter Hausmann, Swiss Re's Head of Cat Perils Europe, location-based data was considered important, especially for flood models, but there are quality problems. The challenges of dynamic data, footprint analysis, and accumulation control were also highlighted.
In terms of future opportunities, telematics was identified as an attractive development as it allows access to real time data. This data could be used to offer additional services to consumers, especially those interested in prevention, accident information, visualisations of their data and more insights into their own risks.
Summary by Brian Rogers, Research Editor, Swiss Re Institute