Modeling firms expand analysis of German floods

The first probabilistic flood models covering all of Germany are being unveiled by modeling firms this month, tools that are expected to bolster insurance underwriting in the region following recent serious flooding.

Risk Management Solutions Inc., of Newark, Calif., plans to announce on September 12, the release of its comprehensive flood model for Germany, which will cover both the risk of flooding from all the major rivers, as well as off-flood plain flooding, the company said.

EQECAT Inc., an Oakland, Calif.-based modeler, says it is coming out with its own model by the end of the month, or early October at the latest. The model expands upon EQECAT's previous model of six river basins, released in 2004, by now covering all postcodes in Germany, as well as all types of flooding.

"It is always beneficial to have new modeling tools to better understand hazards and quantify risk, provided these are used as sensitivity tests on an insurance portfolio and not as pure risk arbitrators," said Andrew Mitchell, business leader, catastrophe management, at Willis Analytics & Solutions, a division of Willis Group Holdings Ltd., in London.

"The availability of two models for a particular country and hazard is especially useful for providing multiple perspectives on the risk, to contrast and analyze the different assumptions used in making computer representations of the physical reality," he said.

"Allianz welcomes any initiative which allows for more transparency concerning hazard assessment," said Olaf Novak, head of catastrophe risk management for German insurer Allianz A.G.'s reinsurance arm. "In this regard, Allianz is already testing these new, upcoming models," he added.

Modelers say the new products are responding to increased demand for risk assessment tools following the major flooding of the Elbe River in 2002. The floods that year caused an insured loss of about e1.8 billion ($2.31 billion) in Germany, and a total economic loss of more than e9 billion ($11.55 billion), the modelers reported.

In terms of the scope of the risk today, if a similar Elbe flood occurred now, it could end up costing e2.5 billion ($3.21 billion) in insured losses, said Michel Voronkoff, the Paris-based manager of European technology and consulting for EQECAT. The increase, he said, is due to inflation, as well more insurance penetration and rising property values.

"I think the 2002 Elbe floods really triggered a very large increase in the market perception of the risk," said Claire Souch, RMS' director of model management.

"In some areas there has been a reluctance to insure," she said.

Flood insurance is generally available in Germany for damage from inland flooding, but not for storm-surge--an onshore gush of water typically caused by a low pressure weather pattern--or ocean flooding, according to insurers.

Overall, the greatest insured risk is along the Rhine River, particularly around Cologne and Frankfurt due to the combination of high flood potential from Germany's largest river and the large concentrations of industrial, commercial and residential property, according to RMS.

By contrast, the risk in Munich is driven predominantly by frequent, smaller flood events, typically caused by rainfall from severe storms and runoff. Hamburg is well protected by flood defenses, but is vulnerable to the largest floods along the River Elbe, which could overwhelm those defenses, says RMS.

RMS' new German flood model will provide probabilistic modeling of the interconnections between different stretches of the same river, different rivers across the country, as well as the off-floodplain events.

EQECAT's comprehensive German flood model will serve as a blueprint for its models in other countries, said Mr. Voronkoff. The model analyzes with a resolution of a few meters.