This project explores the design, implementation, and evaluation of next-generation systems that exploit advances in High Performance Computing (HPC) for Catastrophe Risk Modeling and Reinsurance Analytics.

To better appreciate the details of this project proposal, it is useful to understand the role of reinsurance in today’s economy. Primary insurers issue policies and collect premiums from individuals and businesses against the promise of paying claims when accidents occur. So long as accidents that result in claims are largely uncorrelated, primary insurance companies can use simple actuarial methods and historical claim histories to determine appropriate rates and ensure their own solvency (i.e. capacity to pay) when accidents happen. However, in the case of natural catastrophes, such as earthquakes, hurricanes, flooding etc, there are a huge number of correlated claims and the magnitude of the resulting loss may threaten primary insurers’ capacity to honour their promises. This may trigger not only the insurer’s insolvency, but also financially cripple the insured individuals and businesses who are entitled to insurance even in the face of catastrophic events. It is in this context that reinsurance plays an important role in ensuring the stability and economic benefits of insurance.

Reinsurance companies work in a global marketplace selling (re)insurance to primary insurance companies for slices of risk that the primary insurance companies carry on their books from across many different regions and perils. The reinsurer acts to protect the primary insurers from catastrophic losses so that when big catastrophic events do happen, primary insurance companies with reinsurance protection are unlikely to become insolvent. This allows individuals and businesses to receive the risk protection they have paid for. Thus, reinsurance reduces insolvency risk and strengthens the financial viability of insurance firms by safeguarding capital and reinforcing stability, which ultimately provides insurance customers with assured protection.

At the heart of property catastrophe reinsurance is an analytical pipeline consisting of

1. risk modeling systems to assess and price individual catastrophic risks,
2. risk management tools that can help assess portfolios of catastrophic risk, and
3. risk analysis and optimization tools that integrate catastrophic risk with other forms of risk (including other underwriting risks, investment risk, and macroeconomic risks)

Support for this project is provided in large part by Flagstone Re (www.flagstonere.com), a global reinsurance company that focuses on specialty and property catastrophe reinsurance. Flagstone Re prides itself in its highly technical approach to the reinsurance business. It develops innovative risk analysis and management systems for in-house use in the belief that such systems give its underwriters a proprietary view of natural catastrophic risks and a wealth of germane technical data to support a data driven decision process.

Publications