How to Rank a Location’s Exposure, Hazard, and Vulnerability

As an insurer, you might want to compare relative risk across countries and perils in a region so that you know where you might need to adjust rates or where the best opportunities to develop new businesses lie. Here we demonstrate such an analysis using AIR industry exposures for countries in Southeast Asia.

This region is prone to many natural hazards, including earthquakes and typhoons. One reason is that much of the region is located on the circum-Pacific belt—the “Ring of Fire”—along which 81% of the world's largest earthquakes occur. Another reason is that the frequency and severity of tropical cyclones in the North Pacific basin are greater than the world’s other basins.¹ In fact, on average, more than 25 tropical cyclones form in this region every year and an average of eight or nine typhoons make landfall, one of which is likely to be Category 4 or 5.

Southeast Asia is also a region with a huge concentration of population and economic value. The Metro Manila area and the province of West Java each represent 30–40% of the total insurable value within the Philippines and Indonesia, respectively. Hong Kong and Singapore are two of the most densely populated cities in the world, and their GDP per capita is among the highest. Because of the hazards they face and their major concentrations of exposure these metropolitan areas represent a huge risk to the insurance community.

An Analysis

To make a risk comparison, we investigated all the perils in the countries that AIR models in the region. AIR’s new and updated earthquake and typhoon models for Southeast Asia cover 12 countries in the region, together representing 96% of the region’s total GDP. If we exclude Thailand, Malaysia, and Brunei—which are not supported in the AIR Industry Exposure Database (IED) for Southeast Asia—AIR can assess the insurance risk to territories representing 76% of the total regional GDP.²

There are nine countries for which we have an IED available.³ Of these, AIR offers both earthquake and typhoon models for Hong Kong, Macau, the Philippines, Taiwan, and Vietnam; earthquake models for Indonesia and Singapore; and typhoon models for Guam and Saipan (Table 1).

We ran an analysis for all perils in Touchstone^® based on our IED and calculated the relative risk for AAL, and for the 250- and 1,000-year return periods. The results (Table 2) show that tropical cyclone (TC) in the Philippines is the most costly peril in any country within the region, both in terms of average annual loss (AAL) and extreme losses. The loss ratio (relative risk) is actually more than twice that for earthquake (EQ) in the Philippines on AAL level.

If we exclude Saipan and Guam—which are islands with relatively little exposure—and rank the list, the five top countries and perils in terms of AAL and 250-year return period will be Philippines TC, Philippines EQ, Taiwan TC, Taiwan EQ, and Indonesia EQ. Their rankings for AAL, and for the 250-year and 1,000-year return periods vary, but in all but one case they occupy the top five places in each column. For example Taiwan TC on an AAL basis is higher than Taiwan EQ, but at higher return periods, earthquake risk takes over. Typhoon losses are generally more frequent and smaller, but the rarer large earthquakes have the potential to cause much larger losses.

Vietnam’s earthquake risk also requires our attention: Even though it is not in the top five for AAL, it does enter the top five at the 1,000-year return period, where it ranks even higher than Indonesia earthquake. This means that, although on average the earthquake risk is not high in Vietnam, due to its concentration of exposure, liquefaction risk, and poor building code standards and enforcement, major losses would ensue should an extreme event occur—see my earlier blog on earthquake risk in Vietnam.

This exercise demonstrates the value of relative risk analysis, because it is more than simply a comparison of hazard levels. It is important to remember that, in addition to hazard, exposure and the vulnerability of that exposure determines how much risk a location faces.

Macau and Vietnam are at the lower end of our AAL ranking, in terms of relative risk, while Singapore and Hong Kong are ranked the lowest.

Using the same methods, you can aggregate your exposures to a much higher resolution. Maybe you can compare Manila and Jakarta, Taipei and Hanoi, or selected locations in your portfolio; such analysis could help you with your next underwriting and risk management decisions.

Read “A Multi-Peril View of Catastrophe Risk in China” to learn how the growing value of exposures reinforce the need to take a multi-peril view of risk.