Climate change is no longer an ‘emerging risk.’ It has become a reality, with concentrations of carbon dioxide in the atmosphere now in excess of 400 parts per million and growing evidence of a rise in the frequency and intensity of extreme weather events. But it is more than just the climate that is changing. Rapid urbanisation, ageing infrastructure, increased industrialisation and intricately interconnected factors mean the cost of natural catastrophes is soaring.
Economic losses from natural catastrophes rose to $306bn in 2017, largely driven by the Atlantic Hurricanes Harvey, Irma and Maria. The ten-year moving average for such losses was $190bn, almost double the level seen in the decade ending in 2000.
It is therefore crucial that organisations develop a climate resilience strategy and act on it. To do this, they must first determine the magnitude of the risks they face, i.e. assess the potential scenarios and corresponding financial impact. The second step is to develop mitigation strategies based on these scenarios.
Step 1 – determine the magnitude of risk
Catastrophe modelling software can give a solid understanding of the potential economic impact and relative likelihood of natural catastrophe risks. This type of analysis is only as good as the data that underpins it, so the data must be meticulously checked and verified. Special attention needs to be given to location (preferably latitude and longitude, rather than address) and age of the site.
For those regions and perils not covered by existing catastrophe models, a hazard assessment applying a deterministic scenario-based approach using global hazard maps may be used.
However, one of the problems inherent in tools such as catastrophe models and hazard maps is that they cannot keep pace with the changes in the climate and environment. The evolution of climate change is extremely difficult to model, due to the complexity of the mathematical models and the high degree of interconnection between the variables.
Sustainability programmes to address climate changes, such as green energy or carbon capture, add an additional layer of complexity when modelling climate risks. Historical data is limited in depth and more data is needed to determine the impact of all these changes on catastrophe models.
Another aspect to consider is that catastrophe models do not cover all perils and countries. Other tools, such as global peril-specific hazard maps, are necessary to assess these ‘non-modelled’ perils and regions. Such tools are not as sophisticated as cat modelling software, as they do not include all the parameters necessary to accurately represent the location-specific variations of a specific peril, which could dramatically change within a short distance, for example, effects of soil properties on earthquake shaking levels, or changes of topography within a short distance on flood depths. But they are an essential tool for performing a preliminary analysis of multiple locations with a global footprint to identify the natural hazard exposure levels and, correspondingly, the critical regions and single locations. Experience and judgement, in terms of local topographic conditions, construction practices, or local protection mechanisms, play an important role in analysing the output of the conventional tools used for multi-location hazard identification and assessment.
Besides information pertaining to accumulated annual loss, ‘exceedance’ occurrence probability and other parameters used in the design of the insurance policy, these tools may also help identify high-risk single locations, as well as concentrations of locations that could potentially be affected by a single event.
Prioritisation of locations for the second step of the resilience strategy is based on the definition of ‘critical’ in the organisation. For example, this may be a location or region that meets one or more of the following criteria:
– High concentration of value at one location
– Long replacement time for equipment or stock at a location
– The location is a significant contributor to the group value chain or revenue
– Large concentration of occupants
– Large area around the site that could be impacted environmentally
– Multiple locations that could be affected by a single event.
This review and analysis pertains to operations or locations within the stakeholder’s own responsibility. Ideally, suppliers and critical infrastructure would also be included in the analysis.
Step 2 – develop a mitigation strategy involving insurance and resilience
For those locations defined in the first step as at risk, a deterministic scenario-based loss estimate should be developed, based on detailed information regarding site vulnerabilities (physical and organisational). Such an analysis, which would include an onsite assessment of the reliability and effectiveness of emergency response and business continuity plans, any peril-specific protection measures (seismic gas shut-off values, mobile flood protection elements, etc), quality of structures, and so on, is an essential component of the resilience strategy. With this information in hand, a medium- to long-term resilience strategy can be developed in which budget for capital expenditure projects, as well as reallocation of existing budget toward resilience measures, can be defined.
This type of integrated approach involves not only insurance, which supports the site in restoring operations after the event, but also physical and organisational measures that reduce the impact and severity of an event on the locations.
Contributed by Amar Rahman, risk engineering global practice leader, natural hazards, commercial insurance, Zurich Insurance Company