Stronger analytics for better financial resilience against climate shocks and disasters

• Catastrophe risk models use a bottom-up approach based on scientific and engineering knowledge of hazard, exposure, and vulnerability (each captured explicitly), and the overall model is calibrated and validated based on historical events. This differs from, e.g., economic damage functions based on empirical analysis on historical damages, which do not analyze the underlying hazard, exposure, and vulnerability dynamics driving the risk in detail.
• Catastrophe risk models yield a distribution of damage or financial loss estimates. This can be used as input for, e.g., estimating climate-related contingent liabilities, financial protection gap analysis, and climate risk finance instrument structuring and pricing. It can also be used in modeling chains, e.g., as input for macroeconomic models, the outputs of which can then be used in, for instance, sovereign credit rating analysis.
• There is potential to incorporate future climate projection data into hazard modules, enabling climate-conditioned catastrophe models that can be used to understand potential future climate-related disaster risks.
• Where existing models are unavailable or not suitable, a sequential approach of leveraging existing global and local datasets to provide initial risk estimates prior to investing in the development of a full catastrophe risk model may be appropriate.
• Quantitative analysis is complemented with broader considerations, including basis risk, product quality, additional benefits, and political economy and practical considerations.
• The World Bank has a Climate and Disaster Risk Finance (CDRF) value-for-money process to help governments with their CDRF strategy. Its four steps are: (1) identification of the funding gap, needs, and objectives, (2) assessment of existing and potential instruments, (3) consideration of climate and disaster risk financing strategies, and (4) comparison of suitable products.

Ongoing efforts include better representation of potential climate adaptation measures in catastrophe models, and to extend the models’ analysis of indirect impacts of disasters (e.g., via critical infrastructure and supply chain disruptions). Continuous engagement with experts is also needed to refine hazard and vulnerability modules, and collecting local data for model calibration and validation can help refine vulnerability assumptions to better reflect local circumstances and reduce reliance on assumptions based on data from other countries. Linkages between models can be improved, for example with respect to how information on the geographic distribution of impacts modeled by spatially explicit catastrophe models is retained in macroeconomic and macrofinancial models.