Guy Carpenter expands coverage of EU probabilistic hail model

Reinsurance broker Guy Carpenter, has completed the final development phase of its comprehensive, pioneering probabilistic hail model, the G-CAT Hail Model. The model now boasts enhanced capabilities and has expanded its global reach to include a further seven countries.

Launched in 2014 the G-CAT Hail Model was made available for Austria, Germany, Italy, Poland, and Slovenia, utilising sophisticated and developed software and systems to provide regions vulnerable, and re/insurers exposed to hail events with a more comprehensive view of the peril and its potential threats.

Roughly a year after its release GC has announced further developments with the model, describing the model’s evolution as the “completion of the final phase in the development.”

European hail risks are increasingly part of the insurance-linked securities (ILS) market, as ILS fund managers allocate capital to European catastrophe reinsurance programmes on a collateralized basis, so gaining exposure to this peril.

One such change is the inclusion of a further seven European countries that the G-CAT Hail Model is now capable of covering. This includes Belgium, Czech Republic, France, Luxembourg, the Netherlands, Spain and Switzerland, taking the total number of EU countries that it covers to 12.

Max Strasser, Project Leader for the G-CAT Hail Model commented; “Hailstorm events in Europe have accounted for some of the largest losses in the region in recent years. Most recently, we witnessed the impact of hail and windstorm Ela which affected a number of countries, including Germany, France, Belgium and Austria, and generated hailstones in excess of seven centimetres in diameter (2.8 inches).

“By providing this pioneering probabilistic hail model and extending its geographical reach we hope to greatly expand our understanding of these catastrophic events.”

In addition to expanding the global reach to a further seven regions in the EU, the improved hail model also features an enhanced event set generation methodology.

GC’s Head of Catastrophe Model Development, Mark Weatherhead explains; “The new technique is based on the hail footprints developed from nine years of historic lightning data. This new event set accurately reproduces hail event frequency and intensity by month shown by historic data and is corroborated by evidence from scientific literature on the subject.”

Furthermore, the extensive hail database that supports the G-CAT Hail Model has been improved following the incorporation of additional claims data and information from an increased number of insurers across the modelled regions.

Discussed at the time of launch, and again highlighted by GC in its recent press release on the model’s enhancements, the G-CAT Hail Model’s hazard module adopts a distinctive approach to producing past hail tracks, using an international lightening detection system, known as LINET.

The LINET system works with algorithms produced by nowcast GmbH, an atmospheric research company, explains GC.

In recent times hail events have caused widespread devastation across various EU territories, including the $2 billion insured loss event, wind and hailstorm Ela, which impacted parts of Northern Europe last year.

However, hail events are inherently extremely complex and difficult to asses, model and predict, with experts previously warning of the limitations with experience-based predictions for hailstorms.

Owing to the fundamental challenges surrounding the modelling of hail events, advances in models like the G-CAT Hail Model are likely to prove invaluable to insurance, reinsurance, insurance-linked-securities (ILS) and catastrophe bond market participants.

GC continues to explain that the model produces stochastic event footprints that are enhanced through historical hail footprint scenarios for various recent, large hail events. Each such event is then verified through a comparison with hail event footprints from hail claims data produced by a host of insurers.

To conclude, Matthew Eagle, Head of International Analytics at GC said; “Our G-CAT Hail Model offers our clients a unique means to identify their potential accumulations of insured risks exposed to hail at both a country-level as well as across their entire European portfolio.

“The advances in the model demonstrate once again how Guy Carpenter is constantly seeking ways to expand the boundaries of what is possible in terms of modelling capabilities and enhancing the level of service delivery to its clients.”