The headlines are relentless, loudly proclaiming that climate-fueled extreme weather has caused an insurance crisis, reflected in dramatic rate increases for homeowners and businesses. Some warn that total economic collapse may soon follow.
But as is often the case when it comes to apocalyptic warnings related to climate change, real-world data doesn’t support the narrative.
In reality, the insurance industry, which provides coverage related to hurricanes, fires and other extreme events, is enjoying a streak of record profits.
Defenders of high premiums say it’s because it’s much more expensive to insure homes because of climate change.
But the recent spike in insurance prices is much more likely due, in significant part, to political requirements across the industry that financial companies consider “climate risk,” and the corresponding suite of risk modelers established to meet the newly created demand.
Premiums upfront
In 2009, Warren Buffett of Berkshire Hathaway explained how property/casualty insurers made money: “Insurers receive premiums upfront and pay claims later.” The accumulated premiums, which Buffett called “float,” result in a pile of money that companies invest to earn profits.
Buffett explained that because of vigorous competition among insurance companies, most do not make money from underwriting, they just seek to break even so they can then capitalize on the “float.”
That was then.
In 2024 and 2025, insurance companies made significant profits from underwriting alone. According to the National Association of Insurance Commissioners (NAIC) in a report covering the first six months of 2025, “Despite heavy catastrophe losses, including the costliest wildfires on record, the US Property & Casualty (P&C) industry recorded its best midyear underwriting gain in nearly 20 years.”
In the second half of 2025, things got even better, thanks in large part to hurricanes missing the United States for the first time in a decade.
S&P Market Intelligence announced of third quarter results, “For US P/C insurers, it just doesn’t get any better than this . . . the US property/casualty insurance industry had its best quarter in at least a quarter of a century—and maybe longer.”
The industry’s bountiful financial results of 2025 follow its most profitable year in at least a decade in 2024, according to the NAIC.
But these higher premiums are necessary because insurance companies are paying out more money, right?
Estimating risk
Not always. While insurers set rates based on what they actually have had to pay out, they also rely on forward-looking estimates of risk, typically derived from risk models. Since about 1990, these sophisticated models have supported ratemaking, but recently insurers have been tasked with factoring in climate change to estimates of risk. Starting about a decade ago, the industry’s regulators began raising alarm about the possible effects of changes in climate on banking and finance.
For instance, in 2015, Mark Carney (then the governor of the Bank of England and today the prime minister of Canada) warned that risk experts in the insurance industry might be getting everything wrong: “Currently modeled losses could be undervalued by as much as 50% if recent weather trends were to prove representative of the new normal.”
Such concerns led to new requirements for “climate risk” assessment and disclosure by insurance companies, banks and others in finance.
These requirements resulted in the creation of a new cottage industry — “climate risk” vendors who promised the ability to produce computer models that accurately quantify the effects of climate change on extreme weather and risks of financial loss faced by individual properties.
Yet the science behind such bold promises has been called into question. For instance, one climate scientist warned, “A lot of these bold, hyperlocal claims are greatly outpacing the science.” A model vendor warned similarly, “It’s a Wild West right now.”
Such concerns have been validated by a new study of 13 different climate risk vendors undertaken by the Global Association of Risk Professionals (GARP) on behalf of the Climate Financial Risk Forum.
Wide range of results
The study looked at each of the vendors modeled results for 100 properties around the world, for a range of different extreme events, like floods and windstorms.
The results of the GARP study should send shock waves through the industry and its regulators.
Not only do the vendors not agree with one another, but they also produce an extremely wide range of results. For instance, for a 200-year flood event, some vendors conclude that properties will suffer total losses. For the exact same event and properties, other vendors conclude that there would be no impact whatsoever.
The differences across “climate risk” estimates are huge.
The implications are huge as well. In the face of different risk estimates, academic research argues that risk-averse insurance companies will set their prices at the level of the most extreme estimate, thereby conservatively encompassing all estimates.
If so, that means that increasing insurance rates would be the result of climate change regulations, and not actual changes in climate.
As far as the actual effects of changes in climate on expected annual losses in the insurance industry, Verisk, a catastrophe modeling firm long pre-dating the “climate risk” industry, estimates an annual impact of just 1%.
Insurance companies have spent many decades estimating risk. Perhaps regulators should allow them to come to their own conclusions, rather than insisting they use dodgy science and charge customers even more.
Roger Pielke Jr. is a senior fellow at the American Enterprise Institute who writes at The Honest Broker on Substack.
Insurance companies are making record profits off climate change panic, not facts
