Catastrophe modellers and actuaries at some insurers are being leaned on to trim their climate projections so deals keep flowing. According to anonymous claims from a group of current and recently departed senior executives from across the UK and European non-life, reinsurance and broking markets.
Their memo, Sounding the Alarm, was published on 14 October by Inside Track, a UK non-profit that lets senior staff raise concerns about their industries without being named. It says modellers in some firms have been asked to apply a “haircut” to forward-looking estimates, or to “manage down the numbers”, so as not to cause alarm. The authors say model adjustments are often constrained by commercial considerations, especially where higher physical-risk estimates would hit profitability or cost market share.
That is an awkward charge for a sector whose whole pitch is that it understands risk better than anyone. The insiders are equally blunt about what they think happens next.
The memo sets out what it calls the industry’s default response to rising physical risk, in three stages.
The first is price. Premiums go up to reflect the higher risk and the extra uncertainty. Less affluent households get priced out, and the cost burden rises for sectors such as farming and shipping.
The second is narrower cover. Perils such as wildfire are excluded from buildings policies. Because most cover renews every 12 months, the authors note, insurers can make that change with little friction. They say it is critical that customers actually understand it is happening.
The third is withdrawal. Where premiums and claims can’t be brought back into balance, insurers leave certain geographies and sectors altogether. The memo names coastal, mountain and flood-prone communities as the most exposed.
The authors stress there is no malice in any of this. It is, they argue, simply what short-term incentives reward. The process “has already begun and will now accelerate”, they write, and the net result will be more uninsured losses landing on someone else’s balance sheet. That means households, businesses and, ultimately, taxpayers.
According to the memo, climate barely makes the top three issues in many boardrooms. The reason, it says, is that firms are confident this playbook protects their own exposure.
Read next: Triple-I, Munich Re flag $424 billion protection gap in sweeping risk survey
The section most likely to get attention outside the market is about mortgages. Lenders require buildings insurance. If domestic and commercial property cover is limited or withdrawn at scale, the authors foresee an economic shock that could precede a financial crash, much as sub-prime lending did before the global financial crisis. They argue these are systemic economic risks, not problems confined to a few unlucky postcodes.
Even before cover disappears, they expect higher premiums for food producers, property owners and tourism businesses to feed through into inflation, adding to cost-of-living pressures.
Lord Stern, chair of the Grantham Research Institute on Climate Change and the Environment at the London School of Economics, contributed a commentary to accompany the memo. He argued that the professionals who price risk for a living are now warning, from inside the industry, that their models have underestimated the danger.
The UK is particularly susceptible for this kind of warning. Aviva chief executive Dame Amanda Blanc said recently that England was “for sure” building homes that could one day be uninsurable. Aviva analysed nearly 400,000 new homes and found that 11% already sit in medium or high flood-risk areas. Its earlier analysis of homes built between 2013 and 2022 put the figure at 8%. If current building patterns hold, the insurer expects around 115,000 more homes to go up in flood zones over the next decade.
Read next: One in nine new English homes built in flood zones, Aviva warns
Those new builds get no help from Flood Re, which excludes homes built since 2009. The scheme is also set in law to close in 2039, after which household flood cover moves to fully risk-reflective pricing. Flood Re has itself said that a smooth handover depends on reducing underlying flood risk much faster than is happening today.
Read next: Flood Re warns 2039 transition hinges on faster risk reduction
The European picture is starker still. The European Insurance and Occupational Pensions Authority estimates that only around a quarter of losses from extreme weather and climate-related events across Europe are insured. Allianz board member Günther Thallinger has repeatedly warned that some regions and perils are moving beyond what insurers can cover at affordable prices.
The memo is also critical of the industry’s analytical plumbing. The authors say many firms have world-class capability. Even so, some insurers still rely on inadequate or outdated nat cat models, and the market as a whole leans on a small number of base-case vendor models.
Internal models are expensive to build and good external data is costly. As a result, the memo says, some carriers can’t properly model complex scenarios or aggregate risk across the firm. It also argues that systemic and concentration risk is poorly handled at industry level, because it falls between the silos most firms are organised around.
Some of these points echo concerns already raised by UK supervisors. In its 2025 consultation on climate risk, the Prudential Regulation Authority noted that insurers’ valuations of assets and liabilities were not always consistent in how they reflected climate risk. Its final supervisory statement SS5/25, published in December 2025, sets expectations on governance, scenario analysis and data.
Read next: Insurance associations call for action on widening climate risk gap
The insiders aim part of their message at capital providers. Pension funds that need to pay members more than five or ten years from now should interrogate their exposure to the insurance market, the memo says. In the authors’ view, executive pay is increasingly tied to short-term share price performance rather than long-term value. The industry, they write, is “jeopardising longer-term shareholder value in favour of quarterly growth”.
They also question whether insurers can reliably time their exit from deteriorating risks in a less predictable world. If they get it wrong, the losses could fall on the industry itself.
The memo’s closing argument is pragmatic rather than moral. Even setting aside any wider responsibility, the authors say, insurers have a self-interest in helping to reduce risk rather than simply pricing it and timing their way around it. Otherwise many of today’s policyholders “may suddenly find that they are not our customers tomorrow.”
