Natural Catastrophe Risk Insurance Mechanisms ON THE BUILDING BLOCKS OF CATASTROPHE for Asia and the Pacific INSURANCE 4–5 November 2008 — With Relevance to the Asia-Pacific Region Tokyo, Japan Conference supported by the Asian Development Bank Robert Muir-Wood1 and the Ministry of Finance, Government of Japan Risk Management Solutions Ltd. S U M M A R Y ◉ Catastrophe “CAT” models are fundamental to catastrophe risk management including risk pricing, risk transfer structuring, insurer capital adequacy and risk securitization (either through parametric or loss based triggers). ◉ The CAT model sets out to create a “universe” of all possible events along with their areas of impact (footprint), and the vulnerability of property, people or other assets in the path of the catastrophe. ◉ The standard output of the CAT model is the exceedance probability “EP” relationship (showing the annual likelihood of a loss being in excess of some magnitude) and its integral—the “average annualized loss” or technical rate for the risk. ◉ The development of CAT models has become increasingly technically sophisticated. However, while the essential architecture of the model is simple—and there are open source CAT models—the challenges come through relating the localized hazards to the specific vulnerabilities. ◉ CAT model results are extremely sensitive which is why it is easy to build a bad model. For example, a 1% variation in windspeed (in a hurricane model) makes a 7% variation in loss. A 1m difference in elevation of a property may typically make a 50% difference in the technical rate for flood risk. In developing a new commercial CAT model, the period of research and iterative calibration takes much longer than actually assembling the model. ◉ While a horizontal resolution of 1km–5km might be appropriate for an earthquake CAT model (except in relation to landslide and liquefaction risks), a flood CAT model will require horizontal resolution of 100m or better and vertical resolution no worse than 50cm. ◉ For the Asia Pacific region, flood is the principal hazard, along with earthquake and typhoon wind. In modeling typhoon impacts, it is important to model the wind, storm surge and inland flood perils separately as they will have very different microzonation characteristics. ◉ For modeling the risk to megacities, high resolution data is required for land elevations, the locations and vulnerabilities of all properties and infrastructure, river channels and flood defenses, as well as the characteristics of catchment run-off (and for coastal cities—storm surges). ◉ In expanding the creation of CAT models for the developing world, it is important not to compromise on model resolution or the quality of the research and calibration. From the widespread availability of Digital Terrain Models and the creation of GoogleEarth—there are many new tools that can assist in efficiently developing good quality models. ◉ Given the lack of any agency that vets the quality of CAT models, the transparency of the Florida Hurricane Commission process remains the only way in which to check how different modelers build and calibrate their models. 1 The views expressed in this paper are the views of the authors and do not necessarily reflect the views or policies of the Asian Development Bank (ADB), or its Board of Governors, or the governments they represent. ADB does not guarantee the accuracy of the data included in this paper and accepts no responsibility for any consequence of their use. Terminology used may not necessarily be consistent with ADB official terms.