Effect of earthquake sequences on risk-based catastrophe bond pricing

Catastrophe bonds (cat bond in short) are an alternative risk-transfer instrument used to transfer peril-specific financial risk from governments, financial institutions, or (re)insurers, to the capital market. Current approaches for cat bond pricing are calibrated on seismic mainshocks, and thus do not account for potential effects induced by earthquake sequences. This simplifying assumption implies that damage arises from mainshocks only, while aftershocks yield no damage. Postearthquake field surveys reveal that this assumption is inaccurate. For example, in the 2011 Christchurch Earthquake sequence and 2016-2017 Central Italy Earthquake sequence, aftershocks were responsible for higher economic losses when compared to those caused by mainshocks. This article proposes a time-dependent aggregate loss model that takes into account seismicity clustering and damage accumulation effects in the computation of damage. The model is calibrated on the seismic events recorded during the recent 2016-2017 Central Italy Earthquake sequence. Furthermore, the effects of earthquake sequence on cat bond pricing is explored by implementing the proposed model on five Italian municipalities. The investigation showed that neglecting time-dependency may lead to higher difference (up to 45%) in the cat bond price when compared to standard approaches.

Pricing risk-based catastrophe bonds for earthquakes at an urban scale

Catastrophe risk-based bonds are used by governments, financial institutions and (re)insurers to transfer the financial risk associated to the occurrence of catastrophic events, such as earthquakes, to the capital market. In this study, we show how municipalities prone to earthquakes can use this type of insurance-linked security to protect their building stock and communities from economic losses, and ultimately increase their earthquake resilience. We consider Benevento, a middle-sized historical town in southern Italy, as a case study, although the same approach is applicable to other urban areas in seismically active regions. One of the crucial steps in pricing catastrophe bonds is the computation of aggregate losses. We compute direct economic losses for each exposed asset based on high spatial resolution hazard and exposure models. Finally, we use the simulated loss data to price two types of catastrophe bonds (zero-coupon and coupon bonds) for different thresholds and maturity times. Although the present application focuses on earthquakes, the framework can potentially be applied to other natural disasters, such as hurricanes, floods, and other extreme weather events.