Reinsurers have benefited from favorable economic conditions since the market turned in 2023, allowing them to outperform industry catastrophe trends, according to J.P. Morgan’s recent “Love Actuaries” report.
Reinsurance pricing will surge by nearly 30% in 2023, according to Guy Carpenter.
In addition to these significant rate hikes, structural changes within the reinsurance program have also shifted in favor of reinsurers, characterized by significant increases in attachment points of up to 40% on average, the report said.
At the same time, the historical trajectory of catastrophe insured losses shows a clear long-term upward trend, driven by macro factors such as changing climate patterns and inflation.
The rolling 10-year average of catastrophe insured losses has climbed to about $100 billion, a level that has increased significantly in recent years.
Every year, insurance companies and reinsurers make assumptions about the level of catastrophe losses they will experience in a year. Cat modelers including Verisk predict the level of insured catastrophe losses they expect to see in an “average” year, known as average annual losses.
The projected baseline, which covers factors such as tropical cyclones, earthquakes, storms and wildfires, has risen from $81 billion in 2020 to just over $150 billion in 2025, although different scenarios make it difficult to define a “normal” year, the report noted.
While tropical cyclones, including hurricanes, typically cause physical damage, 2011 showed that unexpected sources can also drive up insured losses. A series of events this year, including the Tohoku earthquake, floods in Thailand and earthquakes in New Zealand, have caused insured losses of well over $100 billion.
JPMorgan Chase emphasized: “By examining Verisk’s forecast data for average annual expected catastrophe losses, we can see that the expected level of insured catastrophe losses has increased significantly in recent years. Driven by trends such as inflation, rising catastrophe loss incidence, and urbanization, this level has increased from approximately $80 billion in 2020 to approximately $150 billion in 2025.”
By comparing “expected” catastrophe losses to actual catastrophe losses, the analysis concluded that 2020 and 2023 were broadly in line with expectations, while insured losses in 2021, 2022 and 2024 were above industry-level expectations.
In 2025 alone, catastrophe insured losses are below the expected baseline for that year. While the results were lower than expected, the total still exceeded $100 billion, and the absolute size of the figure remains high compared with historical levels.
By comparing expected industry losses to actual underwritten catastrophe data, the analysis assesses whether recent years have been light-loss, normal or heavy-loss years and benchmarks reinsurer performance against these market baselines.
Active hurricane seasons drove above-average annual industry disaster losses in 2021 and 2022, resulting in more than $120 billion in disaster losses over the two years.
These levels compare with expectations of $87 billion in 2021 and $100 billion in 2022, or 30% above the average for those two years.
European reinsurers have significantly “underperformed” their benchmarks during this period of high losses, the report said. During 2021-22, their actual disaster losses exceeded designated disaster budgets by approximately 37%.
JP Morgan concluded: “Looking at the evidence from 2023-26, using our methodology, industry losses are around 95% of expected levels, but reinsurers’ catastrophe budgets are around 75% of expected levels, showing improved performance since the attachment point shift.
“Reinsurers typically take on tail risk, so in years with lighter insured losses, it’s fair to assume that reinsurers’ performance relative to industry catastrophe benchmarks should outperform the industry catastrophe benchmark and vice versa. But if we look at 2024, a year with higher than expected catastrophe losses, reinsurers’ catastrophe losses are actually below budgeted levels, whereas the benchmark suggests losses should be higher than normal.”
