What happens when the people paid to price danger decide the danger is too large to price? That is the awkward question now hanging over climate risk. A memo published October 14, 2026 by Inside Track x (Re)Insurance, a UK nonprofit that convenes current and recently departed senior executives at UK, Swiss, and EU insurers, reinsurers, and brokers, argues that climate stress is not just an environmental problem. It is a balance-sheet problem that could travel through property cover, mortgages, pensions, and then the wider economy. The warning is not that insurance will vanish overnight. It is more unsettling than that: the system may fail by degrees, like a levee that holds until it does not.
The memo’s value lies less in drama than in sequence. It describes a three-stage insurer response to climate risk: first, premiums rise; then coverage is narrowed by excluding perils such as flood or wildfire; finally, insurers withdraw from geographies or sectors altogether. That progression is easy to miss because each step can be defended as prudent management. Together, though, they create a slow-moving withdrawal of trust. Markets often treat such moves as isolated underwriting decisions. In reality, they are signals that the private sector is re-rating the future faster than borrowers, lenders, and policymakers can absorb.
The memo argues that the danger becomes systemic when property insurance is no longer available at scale. Its bluntest line says: “Insurance is necessary to access a mortgage, allowing providers to manage their own financial exposure. As both domestic and commercial property insurance is limited or withdrawn at a significant scale, we foresee an economic risk that could precede a financial crash in the way that the sub-prime mortgage crisis catalyzed the Global Financial Crisis.” That is not a claim that climate risk equals 2008 in every detail. It is a warning about plumbing. If the pipe from insurance to lending narrows, mortgages do not disappear gracefully; they become harder to originate, refinance, or roll over.
A second quote in the memo sharpens the point: “The undermining of the mortgage market drove the Global Financial Crisis – we are worried a major increase in uninsured, high-risk properties could manifest in a similar way as climate impacts intensify.” Here the comparison to subprime is useful precisely because it is imperfect. The last crisis was not caused by a single bad asset class; it was caused by a chain of incentives, ratings, leverage, and false confidence. Climate retreat could become a cousin to that crisis through a different asset, but the same human habit: assuming a local underwriting problem cannot become a systemwide financing problem until after the fact.
The memo says the effects will be felt globally, including North America, even though it is addressed to UK and European stakeholders. That matters because fragility migrates. A risk that begins as a local pricing issue in insurance can travel into global capital allocation once lenders, pension funds, and commercial property owners react to higher costs and tighter availability. Finance likes to imagine that risks are compartmentalized, but the real world is rarely so polite. In nature, a dry forest does not need every tree to burn to become dangerous; it only needs enough continuity. In markets, continuity often comes from assumptions, not contracts.
The memo also echoes an older institutional problem: the time mismatch between insurance and mortgage finance. Insurance renews annually. Mortgages can run for years. According to the evidence pack, OSFI and Quebec’s AMF had already flagged that mismatch, including the risk that lenders may not know when cover lapses. That is a small administrative detail with large consequences. Systems often fail where monitoring is weakest. If the lender believes the roof is protected and the insurer has already stepped back, the borrower sits between them like a passerby on a bridge with rusted bolts hidden under fresh paint.
The memo goes further by alleging that some firms pressured catastrophe modellers and actuaries to “manage down the numbers,” or apply a “haircut” to projections, to protect profitability and deal flow. Another attributed line says, “Modelling adjustments are frequently constrained by commercial considerations, particularly where higher estimates of future physical risk could result in reduced profitability or loss of market share.” That accusation should be handled carefully. It is not proof that every model is bent, nor that every insurer is acting the same way. But it is a reminder that models live inside institutions, and institutions answer to revenue. A risk estimate is only as sober as the incentives around it.
This is where investor psychology usually goes wrong. People want climate risk to appear as a clean trend with a neat probability attached to it. They want a chart that says when the pain will arrive and how fast it will compound. But systemic risk does not usually announce itself that way. It starts as a weak signal, then a nuisance, then a margin issue, then a funding issue. By the time it looks obvious, it is often too late to enter at a sensible price. That is why the right question is not whether one insurer will lose money in one geography. It is whether the whole structure becomes less financeable as physical risk intensifies.
Lord Nicholas Stern, chair of the Grantham Research Institute at the London School of Economics, endorsed the memo and tied it to his earlier work. He said, “Twenty years ago in the Stern review, I said that the cost of inaction was much greater than the cost of action on climate change… Now the people who price risk for a living are warning, from the inside, that our models have underestimated the danger and that mispriced risk is building to potentially unmanageable levels.” Stern’s point is not that catastrophe is certain on a timetable. It is that inaction can look cheap until the bills arrive in clusters. That is one of the oldest political errors: postponing costs until they reappear as losses elsewhere.
There is also a deeper game-theory problem here. If insurers raise prices first, they preserve capital but strain affordability. If they hold prices down, they may keep market share but absorb losses later. If they retreat, they protect solvency but push risk outward to homeowners, lenders, and the state. Each player is making a locally rational move. Collectively, those moves can produce a worse outcome than any one actor intended. This is the tragedy of distributed caution: each firm protects itself, and the system becomes more brittle.
The memo is framed as a call to policymakers, regulators, and investors, not as a scheduled event with a vote or hearing attached. That makes the warning harder to trade, which is precisely why it matters. Markets are often misled by the absence of a date. They assume no calendar means no urgency. But fragility does not respect event schedules. The FCA had already published a web page on physical climate risk to property insurance and mortgage markets, including Flood Re’s scheduled 2039 expiry, a reminder that some of the system’s relief valves already have end dates. Long before that point, the market may reprice the burden in ways that ordinary borrowers cannot easily see.
The most uncomfortable implication is that “insurable” may become a moving target. Once certain properties or regions are deemed too dangerous or too expensive to cover, access to credit becomes a proxy for geography and climate exposure. That is not merely a housing issue. It is a capital-allocation mechanism. The market will call it pricing discipline. Families will call it exclusion. Regulators will call it a challenge. History will call it predictable.
There is a familiar pattern in finance: a risk is discounted because it is diffuse, then dismissed because it is inconvenient, then finally discovered after it has metastasized into leverage, illiquidity, or both. Climate retreat by insurers fits that pattern too well. The memo does not prove an imminent crash. It does something more valuable. It identifies a brittle joint in the machine, where annual insurance meets long-duration debt and where commercial pressure can quietly distort risk models. That is how modern systems break: not with a scream, but with a paperwork chain that no one wanted to inspect.
The lesson for investors is not to chase a headline or declare the end of property finance. It is to stop confusing price with safety. A premium that looks manageable today may simply be the market’s way of delaying recognition. In classical terms, the wise builder does not wait for the earthquake to test the beam. In market terms, the prudent observer asks where the load is transferred when one support gives way. On climate risk, that support may already be shifting from insurers to homeowners, then from homeowners to lenders, and from lenders to the public balance sheet. That is not a theory. It is how fragility behaves when everyone calls it prudent.
