RMS Adds Inland Flood Peril to China Catastrophe Model Suite: Now Models 99 Percent of Property Average Annual Loss

China Flood Insurance: Market Challenges and Opportunities

China is confronting a future where the frequency and severity of flood events are increasing, influenced by a range of factors such as climate change, together with exposure becoming much more concentrated; 80 percent of the population is projected to live in cities by 2050. As a vast country, flood risk is low across the deserts in the north and west, while the hilly regions of central and southern China are prone to flash floods. The densely populated eastern plains, including the Yangtze and Yellow River deltas, experience the greatest flood risk.

The country is not standing still. Between 1998 and 2017, China invested US$294 billion in flood defenses, and decreased the ratio of economic flood loss to GDP from 2.3 percent in the 1990s to 0.4 percent in the 2010s. Insurance penetration remains low and decreases when moving inland away from the typhoon-exposed eastern coastal areas, even though fluvial and pluvial risks remain.

There are several large cities and clusters of exposure further inland, often sited along major rivers that have experienced regular floods during the course of history. Nonetheless, aspects such as increasing exposure values and huge investments in flood defenses, China represents an attractive, growing market for insurers who understand the rising flood risk.

As both a secondary peril and a sub-peril to typhoon in China, flood presents unique challenges to (re)insurers. Insurers covering China need risk modeling that captures both typhoon and inland flood, all covered within “wet peril” policies, typical of those written in China. It’s easy to assume that the main risk driver comes from tropical cyclone flooding.

In fact, around 70 percent of “wet perils” economic losses come from non-tropical-cyclone-driven flooding, from sources including monsoon rains, heavy rainstorms, or snowmelt. Insurers leave themselves exposed if they do not model non-tropical-cyclone flooding, underscoring the importance of combining both models to get a full picture of “wet perils” risk.

The rapid development of the Chinese economy means that insurers writing flood coverage must also account for non-climatic factors and understand changes to exposure linked to urbanization and land use in order to make sense of the risk.

There is a large protection gap for flood in China, insured flood losses usually account for only about two percent of economic losses. For (re)insurers who are committed to China and are building their presence for the long-term, there are sizeable opportunities in the region – using modeling that can fully evaluate the risk.

Introducing the Most Comprehensive View of Flood Risk in China

The RMS® China Inland Flood Model covers the entirety of mainland China, including Hong Kong and Macau, and is based on 10,000 years of simulated precipitation and 100,000 simulated flood events. Calibrated against 20 years of recent observed data, the model captures both fluvial and pluvial flooding, along with the spatial correlation of flood risk. This includes simulating over 355,000 kilometers (220,568 miles) of river network—more than our Europe and India Flood Models combined—and modeling 640 reservoirs and lakes.

River discharge is calibrated using data from 1,600 stations throughout China, with over 6 million data points digitized from hard-copy reports and manually quality assessed. The model also works together with the RMS China Typhoon Model, for a combined view of “wet perils” risk.

Building the Most Realistic View of Risk with Strong Local Partnerships

To ensure model quality, RMS partners with more local agencies, institutions, and subject matter experts than any other vendor, to incorporate the best-available local intelligence into our models. The modeled flood hazard footprints were validated against 350 flood hazard maps from local agencies.

Recognizing that mitigation and flood defenses can make a huge difference in flood risk, RMS also partnered with Beijing Normal University to understand flood defense standards of protection in 70 cities across China. With this information, we were able to extrapolate the standard of flood defenses across all regions in the country and better understand exposure in areas with higher risk concentrations.

Finally, event losses are also calibrated and validated using client exposure data which represents over 70 percent of the market, and historical losses from hundreds of flood events spanning 70 years – to realistically reflect the flood risk across China.

Plug the Flood Insurance Gap with Confidence - Even in Low-Resolution Areas

All insurers work hard to improve their exposure data for accurate risk analysis. Flood is a high-gradient peril and often requires a building-by-building risk assessment to achieve profitable risk selection and underwriting. The RMS China Inland Flood Model enables (re)insurers to access high-resolution insights with flood hazard modeled down to 50-meter resolution in high-exposure areas, capturing the high gradient nature of the flood peril.

Capitalizing on rich exposure data, the model has 91 unique primary damage curves for buildings and several secondary modifiers to further refine building vulnerability. It includes a new building inventory that contains information on 60,000 industrial clusters and 2,400 tall buildings, to help fill any gaps in exposure data.

In areas where less exposure data is captured or only aggregate data is available, (re)insurers can lose the fidelity of model results leading to poor risk selection. RMS combines a sophisticated loss-weighted hazard aggregation methodology with modeled losses calibrated at province level, to ensure clients with low-resolution or aggregate data can use the model results with confidence.

Summary

RMS is the first and only vendor to offer peril models for the three main cat risks in China (including Hong Kong and Macau) for earthquake, typhoon, and flood. With the most comprehensive set of model solutions for the market, (re)insurers can now model more than 99 percent of cat risk using a unified and consistent methodology.  

To find out more about the RMS China Inland Flood Model, click here.