As we blogged recently, earthquakes strike around the clock and around the globe, hitting some communities far harder than others. Because of this inequity, helping to close the protection gap for vulnerable countries has become a priority for governments and non-governmental organizations alike.
To this end, Chile, Peru, Colombia, and Mexico have recently issued a parametric catastrophe bond to the Insurance-linked securities (ILS) market, obtaining more than a billion dollars in cover. This is the largest sovereign risk insurance transaction ever and the second largest issuance in the history of the catastrophe bond market. It is the first time that Chile, Colombia, and Peru have accessed the capital markets to obtain insurance for natural disasters. AIR Worldwide was the modeling and calculation agent contributing to the design of this bond for the World Bank Group, the leading provider of natural disaster risk insurance for emerging and developing countries. This is the World Bank's largest catastrophe bond transaction to date.
Catastrophe bonds are mechanisms through which catastrophe risk is transferred from the sponsor (a party or group of parties) to the capital markets. In general, the issuance process begins with the selection of an independent modeling agent (a catastrophe modeling firm such as AIR), legal counsel to assist with regulatory compliance for a securities offering, and a structuring agent (usually an investment bank or the capital markets arm of a major broker or reinsurer) that provides advice on designing and placing the bond. Together, the sponsor and the structuring agent define the set of conditions that will trigger a payment from the investors to the affected country (the trigger).
In this case, AIR Worldwide helped with the design of a cat-in-a-box parametric trigger. This works by dividing the national territories into grids, to each square (box) of which an appropriate magnitude and depth threshold for earthquakes is defined and assigned. Should an earthquake strike, data from a respected third party (here the U.S. Geological Survey) is used to determine if its magnitude was greater than the threshold magnitude of the box in which it occurred. If it is, the cat bond will pay out.
The design of the boxes is guided by a probability of attachment and an expected loss probability. It is not an easy task to arrive at those two target probabilities. AIR and the World Bank worked closely with the issuing countries to define them through financial analysis of their existing disaster finance instruments. Once they are defined, the trigger design becomes an optimization problem.
Let’s say that in an analysis a certain country has one loss causing event per year using AIR’s 10,000-year simulated event catalog—10,000 loss causing events. Let’s also say that the target probability of attachment is 2%, meaning that the trigger can cause a payout for 200 of the 10,000 events. Many sets of 200 events can be triggered by setting various magnitude thresholds; however, each set of 200 events will be different. The goal of the optimization process is to find the one set of 200 events that causes the most loss over the probability of attachment. This trigger increases the odds, as much as possible, that the country will receive a payout should a severe event occur—but will not receive one for events that cause an insignificant loss or for events that cause losses meant to be covered by other instruments or by the private insurance industry.
Much of the work done by AIR Worldwide in this transaction was associated with developing an algorithm capable of quickly performing this optimization and reporting on the risk metrics of the final triggers.
This transaction has as its goal the provision of emergency funds to the member nations should earthquake disasters strike. South America is one of the most seismically active areas of the world. It was therefore a pleasure to support a transaction that ultimately makes the participating nations more resilient.