Extraordinary coastal growth means 2025 could be a unthinkably expensive disaster year

Swiss Re warns there's a 1-in-10 chance this could be a year of $300 billion in disaster payouts -- but we keep building sandcastles on the beach

Any toddler knows—or quickly learns—that if you build a sandcastle on a beach and fill its moat with favorite toys a wave will come along and wreak destruction. Well, think of US coastal communities as magnificent, intricately built networks of sandy structures. For decades, we have been building homes, businesses, and infrastructure, all packed closely together and crammed with treasures. That "value accumulation" leads to very high claims when storms or flooding arrive.

That's why a report by reinsurance giant Swiss Re said earlier this week that insured losses from disasters are trending straight up. Hurricane Ian (2022) cost insurance companies a whopping $64 billion—even though tougher building rules had been put in place following the devastation caused by Hurricane Andrew in 1992. Between 1975 and 2022, the number of people setting up shop in Ian's path—Fort Myers, Cape Coral, and Punta Gorda FL—grew more than sixfold.

The US is an exceptional place when it comes to physical climate risk: In 2024, America accounted for almost 80 percent of global insured losses caused by natural disasters, with about half of these losses happening in Florida, Texas, California, Louisiana, and Colorado—states that have significant development and wealth concentrated in risky areas. In general, if you live in a state that is likely to experience natural disasters, you are highly likely to pay more for insurance—because insurers know that exposures are high and growing.

The risk/insurance picture is truly alarming for Florida—a state whose population has nearly doubled over the last 50 years. Swiss Re researchers keep running computer models of old hurricanes through newly built-up Florida regions, and the outlook is grim: If Hurricane Andrew arrived today it would cause three times the insured losses it did in 1992 because so many more people now live in its path. And the Great Miami Hurricane of 1926, if it arrived this summer, could cause twice as many insured losses compared to re-runs of either Andrew or Katrina. That's why, Swiss Re says, when you look at the ZIP-code level in Florida, high insurance costs are "strongly localized" to coastal communities: in 11 percent of Florida ZIP codes, yearly premiums are over $10K a year; in about 4 percent of ZIP codes, premiums are more than $15K. Those expensive places are coastal ZIP codes.

At this point, if you ignore really big disaster years, the amount of money paid out in America by insurance companies following natural disasters is already going up steadily by about 5 to 7 percent a year, even after adjusting for inflation. But some years, with major, direct disasters (like huge hurricanes), these losses go way up: since 1995, those spiky years have been 1999, 2004, 2005 (Katrina), 2011, and 2017 (Harvey and Maria).Those peak years—roughly one in every six—collectively account for about a third of all the money (about $2.14 trillion) insurers paid out for natural disasters over those three decades.

Swiss Re is saying that peak years are not anomalies. They are now expected. It's been more than six years since the last peak year, but Swiss Re isn't relying on that rule of thumb in predicting what's possible in 2025. Instead, they used dozens of bulked-up computer models that include information about the number and value of buildings along the coast as well as climate data, and emerged with the assessment that there's a 1 in 10 chance that 2025 will be a really, really expensive year for insurance companies—costing them over $300 billion in payouts. Another peak year will mean more exits by those companies, and more risky areas becoming (effectively) uninsurable--or covered only by thin, high-deductible, risky insurance provided by risky insurers.

Even without another peak year this time around, risky areas in Florida will see—inevitably—more declines in property values, more increases in insurance premiums, and more owners giving up their homes as the costs and risks become unsustainable. Prices in Punta Gorda, which was hit hard by Hurricane Helene in 2024, are already down 35 percent.

As a nation, we want insurable properties—necessary for the mortgage market to keep ticking along. We also want to avoid people "going bare," or going without insurance because they can't afford it, because the costs of recovery from the assaults of nature will be borne by everyone in some form. And all of this has to happen against a background of predictably increasing disasters, chronically increasing risks, and understandable insurer skittishness.

What's the right response? Swiss Re says the key is "coordinated efforts across stakeholders," which should include at least three things: (1) better building codes, although that's not a complete answer—remember Hurricane Ian, (2) improved zoning, so that we stop new development in floodplains and floodproof buildings that are already there; and (3) investment in safer infrastructure, which should include steadily decommissioning infrastructure that is unsafe.

In the abstract, and from a safe European distance, that all sounds simple enough. But even changing building codes at the local level in the US is astonishingly difficult.

Last week, I heard a presentation about a well-meaning American Society of Civil Engineers effort to update flood-related building standards—covering how we place and elevate buildings so they won't flood. It's a long process to achieve consensus, and some of the standards hadn't been updated since 1998. After the ASCE does its work, the International Building Code people decide whether they'll incorporate it in their standards, which are updated once every three years. ASCE couldn't get its work into the 2024 IBC document, so now everyone is hoping adoption will happen in 2027. And then local governments, states and cities, decide whether to adopt whatever comes out of that process into their own building codes.

That's when developers and builders put up a buzzsaw of opposition. People who watch coastlines and flooding know that what we used to think of as a "500-year flood," or one that has a 0.2 percent chance of happening every year (a 1 out of 500 chance) is now effectively a "100-year flood," with a 1 percent chance of happening every year and a 26 percent or so chance of happening over the course of a 30-year mortgage. Our nomenclature is outdated given the realities of climate change. But when the engineers say "you shouldn't build a house in a 500-year floodplain without doing XY," the construction crowd balks. They'll say that will make the house more expensive, so fewer people will be able to buy it. No one tells them how much money society will save if the house is safer, and that probably wouldn't change their minds anyway.

Local governments often listen to the builders—we're in a housing crisis, after all—which means we are stuck with retrograde building codes and less-safe houses. We end up with a worse product that more people can buy, but that in the long term will experience devastating flooding. Modest elevation of houses in the 500-year floodplain would save millions, as would retreating from the current 100-year floodplain altogether.

So if Swiss Re really wants to ensure more insurable communities, they and their brethren will need to get involved in the building code world, conditioning the provision of re/insurance on adoption of humane standards when it comes to where and how we live. Someone with money has to make this move.

Left to our own devices, we apparently have no incentive to do anything other than continue building lavish sandcastles on the shore. Any toddler would know better.