Sign in
Catastrophe bonds — a cornerstone of last year’s hedge fund strategy on the back of high returns — are facing challenges due to unstable and short weather patterns, Bloomberg reported. These bonds, along with other ins.-linked securities, rely on risk models that may not fully account for the increasing frequency of climate change driven weather-related events like wildfires and thunderstorms.
What are catastrophe bonds? Catastrophe bonds — also known as cat bonds — are high-yield debt instruments designed to help ins. companies raise money in the event of a natural disaster. They are also frequently used to mitigate the effects of climate change — which can exacerbate and increase the frequency of some natural disasters — by transferring a specified set of risks from a sponsor, typically an ins. company, to investors. Ins. companies don’t have to pay interest or repay the principal they loaned if a disaster results in a payout from the bond.
The star child is on shaky ground: The USD 47 bn catastrophe bond market was at the heart of the highest-returning hedge fund strategy last year with a 20% value increase, offering high returns to investors including Fermat Capital Management, Bloomberg added. However, the smaller weather shocks fueled by climate change are putting these returns at risk, as the models used to build these bonds may not be able to predict the new breed of high-frequency events.
Without accurate prediction, bond prices will not be reflective of real value: For cat bonds to become common and commercialized in a “climate-themed bond universe,” they would need to reflect the risk of extreme weather brought about by climate change for the price to be adjusted accordingly, Climate Bonds Initiatives explains. “Without accurate climate change forecasts factored in, climate-aware investors may actively avoid purchasing cat bonds given that they do not pay out if extreme events occur,” the initiative writes. To align with climate-aware investment strategies, cat bonds will have to incorporate accurate climate change risk models.
Climate change is making it more difficult to model risks: Current methods need improvement to address climate change, according to catastrophe modeling firm Karen Clark. The company updates models every two years, but challenges include modeling complex weather events and predicting future risks as the climate changes. Researchers are looking for better ways to assess these factors to improve the accuracy of catastrophe models.
Especially for smaller catastrophes which are more difficult to predict: Secondary perils — catastrophes that result in small to mid-sized losses, such as hail, flood, storm or bushfires — are now causing the majority of global ins. losses. In the hottest year on record, these perils accounted for 86% of losses. Mid-sized events with damages between USD 1 bn to USD 5 bn are becoming more common, posing a challenge for both ins.ers and investors. Unlike primary perils like hurricanes, with well-documented historical data, secondary perils are more unpredictable. Wildfires and storms often affect wider areas and evolve rapidly, making traditional risk models ineffective.
But some US companies have improved their methods: Catastrophe bond specialist Elementum Advisors revised its wildfire model based on historical trends that no longer match today’s climate reality. After analyzing data from nearly 2 mn US wildfires, Elementum found a higher frequency of areas burned in northern California than the model predicted. This led to more accurate estimates and better negotiation of interest rates on agreements.
Others looked towards AI for accuracy: Catastrophe modeler Verisk Analytics recently revamped their model for severe convective storms — a critical secondary peril — using machine learning to analyze two decades of radar data. The updates allow them to pinpoint “every single point of every single event for 20 years.” The revised model reveals a significant shift in thunderstorm risk, and Verisk predicts climate change will continue to reshape these patterns.
