Insurance markets don't become uninsurable overnight. But the transition from insurable to uninsurable is the final stage of a sequence that can take years or decades to play out.
By adopting green infrastructure in place of some gray infrastructure, Kansas City reduced the total costs of managing its combined sewer system through 2040 by over $2 billion.
Cities spent decades building pipes and pumps to move water out fast. Hoboken tried absorbing it instead—and cut flooding by 88%.
Insurance markets don't become uninsurable overnight. But the transition from insurable to uninsurable is the final stage of a sequence that can take years or decades to play out.
Insurance markets don't become uninsurable overnight. But the transition from insurable to uninsurable is the final stage of a sequence that can take years or decades to play out—a chain reaction of rising hazard exposure, accumulated losses, revised models, reinsurance repricing, shifting corporate strategies, and regulatory friction.
By the time an insurer declines to renew coverage in the Pacific Palisades or Altadena, for example, the forces that have made that market unprofitable have long been in motion.
By going back in time to understand the step-by-step process by which a market becomes uninsurable, regulators, investors, and policymakers can see which assets and regions are approaching the same fate.
Insurers are not public utilities. They make strategic decisions about where to deploy capital based on scale, peril exposure, diversification, and regulatory conditions. Presence in a market, like California or Florida, isn’t an obligation for these companies. California is the fourth-largest insurance market in the world, which has historically made it worth tolerating regulatory constraints and elevated risk.
Francis Bouchard, a contributing editor for The Epicenter and managing director of climate for Marsh McLennan, spent 35 years in the insurance and reinsurance industry. "An insurer in the California market illustrates a desire to write business in California," he says. "They're not automatically there. They're not forced to be there."
Those choices are shaped by geographic strategy, peril appetite, customer segment, and the regulatory environment in which an insurer is willing to operate.
Once in a market, insurers manage portfolios of homes, but the profile of the overall portfolio matters far more than the characteristics of any individual property. Insurers track concentration by county, ZIP code, and peril, constantly monitoring how much of any one risk type they're accumulating.
A well-built, fire-resistant home can still be rejected by an insurer—not because there’s something uninsurable about that specific house, but because the insurer has no more room for that exposure type.
A homeowner might have a fire-resilient home, built to Wildfire Prepared Home Plus standards from the Insurance Institute for Business & Home Safety (IBHS). But the insurer may decline to write a policy because it is already maxed out in California wildfire coverage, and it isn’t seeking more exposure.
When losses force insurers to reassess their exposure, the trigger doesn't need to have occurred in the same market. A major wildfire event in New Mexico, for example, prompts insurers to review their book and recognize they have clients with a similar risk profile in California. A loss event anywhere can shift assumptions everywhere. In California, the 2017 and 2018 wildfire seasons’ $20 billion in losses wiped out more than two decades of accumulated underwriting profit—and the 2025 LA fires generated another $40+ billion in losses.
Profitability erosion triggers a reset. Modeling firms revise their risk estimates and release updated outputs. When those outputs show elevated exposure, insurers have to decide how to react. Smaller insurers, who rely more on third-party models than proprietary analyses, can be the first to make changes.
This reset is a slow process. Modeling firms evaluate new data, generate revised models, and then insurers evaluate the models. Bouchard describes the relationship between insurer and modeling firms as something like interpreting a sacred text: “The modelers claim to have the truth. But the companies will take it and say, ‘I believe you over here, but I want to test this assumption or add that secondary factor.’” Big insurers treat model outputs as one ingredient in a broader view; smaller ones may have less to go on.
Reinsurers can move faster. Operating largely outside state rate regulation, they can tighten terms or raise prices immediately. When reinsurance costs spike, primary insurers are affected immediately, especially smaller insurers that rely more on reinsurance. Once an insurance company’s margins are compressed by rising reinsurance costs, it will want to adjust its own premiums to compensate, leading to Stage Five.
After evaluating the new risk data and potentially bearing higher reinsurance costs, insurers will reprice their premiums to reflect the new information. In heavily regulated states like California, that process has historically taken up to two years, though Commissioner Ricardo Lara has pledged to accelerate the review process. However, this disconnect between fast-moving risk and slow-moving rate approvals can create a structural problem.
"Imagine if that same rule applied to gas stations, which every day are reflecting what's going on in the Strait of Hormuz," Bouchard says. "To change our price, we need government approval, and it takes a long time to get it. You're really hamstrung."
Insurers facing inadequate prices have limited options: absorb losses, reduce exposure through nonrenewal, or exit. Washington state offers a useful contrast—its more responsive regulatory framework has kept its insurer of last resort to a fraction of California's FAIR Plan enrollment. There are 431,000 structures in California enrolled in FAIR. In Washington, there are 361.
As insurers exit or sharply raise prices, a collective judgment forms. Those remaining face an escalation of the problem: The highest-risk policies accumulate on their books as more cautious competitors leave. As a result, profitability deteriorates further. The market doesn't collapse like a bank run. Instead, an insurer becomes progressively less willing to insure at prior terms until it is confident it can deliver on its shareholders’ expectations.
This sequence has played out in California's wildfire insurance market. Now, the same pattern is playing out in severe convective storm markets, where hail-loss data is forcing insurers to rethink their assumptions about risk in the Midwest and South.
Knowing the sequence can help identify the next vulnerable market. To determine which assets are approaching uninsurability, look for the warning signs:
Where those conditions converge, markets becoming uninsurable could be next. "The only way to keep this from happening over and over again," Bouchard says, "is to get our arms around the risk."
First, that means intervening before development locks in risk—in the pre-development stage through building standards, land use decisions, and insurance requirements that reflect actual hazard exposure rather than minimum compliance thresholds. Then, it means applying a resilient lens to construction and insurance—through defensible space, fire-resistant construction, and community-wide mitigation efforts (for more, check out our profile on the Dixon Trail Community in California).
Capital will follow demonstrable, measurable risk reduction. The markets that build that case early are the ones that will stay insurable.
Have thoughts to share on this piece, or want to add your voice to the conversation? Reach out!
The economic losses from disasters that are not covered by insurance continue to grow, but resilience projects are generating measurable positive returns.
If today’s funding landscape resembles disconnected wells, then climate resilience requires something closer to a watershed—where resources are intentionally pooled, directed, and circulated across geographies, sectors, and time horizons.
Climate resilience is often framed as a problem of insufficient funding. In reality, it is a problem of how funding is structured and deployed.
Wildfire-prone communities across the West are no longer waiting for federal dollars to mitigate wildfire risk. As FEMA funding stalls and detection technology gets cheaper, local districts are investing in wildfire-detection sensors, drones, and AI-monitored cameras.
