Modelling climate physical risks

Dapeng Hu, Mark Hu, Amit Madaan and Shubhin Puri

Investment risk associated with physical changes of the world – from rising temperature and sea levels to the increasing frequency and intensity of extreme events such as hurricanes, wildfires and floods – has been well known by the financial industry. Yet, the implications for security selection, portfolios construction and risk management have been notoriously hard for investors to grasp. The effects of slower-moving physical changes such as rising sea levels can seem distant, causing investors to discount pressing climate-related risks already lurking in portfolios; the risks are hard to quantify or even to measure systematically. New climate patterns caused by climate change mean long-dated historical data alone can be a problematic guide to the future, and investors using models overly reliant on the past may miss the long-run climate trend. Further, although drilling down on physical risk to the exact geographical location and asset level is key to investors, analysing huge amounts of climate data properly and effectively has been challenging as it requires a combination of climate science, economics, finance and data science knowledge and tools.

BlackRock has conducted