If floodwaters rise to 8 inches outside a warehouse owned by private investment firm Postlane Partners in Freeport, Texas, an internet-connected sensor would automatically trigger a $3 million payment from an insurance provider, a startup called FloodFlash. If the water level rises to 16 inches, the payment will be increased to $5 million.
This sort of coverage, known as parametric insurance, functions similarly to a bet. It has a fixed payout, which minimizes the insurer’s obligation. It also comes from a new brand of insurance company, which is stepping in to shoulder risks that traditional insurers can’t—or won’t—take on because climate change has rendered more traditional types of coverage unprofitable.
Owners of homes and businesses have watched with alarm as major insurance companies have stopped offering coverage in California, Florida, and other parts of the country prone to natural disasters. But this shift has also created an opportunity for new types of insurers.
The secret sauce of these startups is technology. They are using better data science and incorporating artificial intelligence. Some, like FloodFlash, use on-the-ground sensors that enable an automatic payout when a catastrophe occurs.
The need for new ways to insure against catastrophes arises from the increasingly extreme nature of our planet’s weather. As we put more infrastructure of every kind into harm’s way, that’s leading to bigger losses for insurers. In the 1980s, the U.S. suffered an extreme weather event that cost $1 billion every four months. Now, one is happening every three weeks, according to the U.S. National Climate Assessment, released in November.
“If you had to pick a canary-in-the-coal-mine industry to measure the extent to which climate change is real, I think insurance is probably the best one I can think of,” says Max Clarke, chief executive of Plover Parametrics, which uses data to structure parametric coverage offered by insurance companies. “The balance sheets—they’re not going to lie.”
Sophisticated reinsurers like Munich Re and Swiss Re have long used the kinds of data science some of these startups employ. But reinsurers are mostly in the business of backstopping the kinds of retail insurers from which consumers typically buy policies. What’s new is the availability of these kinds of policies for everyday customers.
Insurance startups have an opportunity to grab more customers partly because some regulators are making it harder for traditional insurers to pass on the costs of more expensive and more frequent disasters.
In fire-prone California, for example, property insurance companies aren’t allowed to factor in the rapidly rising cost of their own reinsurance when setting premiums. Until earlier this year, they also weren’t allowed to use projections of future, climate-driven increases in the cost of weather disasters. Regulations like this are among the factors that Allstate and State Farm cited when they announced in the past year they would stop writing new policies in California.
It isn’t clear if possible future regulatory or legislative changes will bring companies back into markets they’ve left—or whether customers will be able to afford their coverage when they do. Climate change isn’t slowing down, which means it’s likely more and more people will be forced to find alternative coverage.
As a result, some homeowners and business owners are seeking out insurance from companies that aren’t subject to the same rules as traditional insurers. These companies are known as “non-admitted” insurers, and they usually charge higher rates for less coverage.
“The big guys are saying, OK, if you won’t change the rules, then I’m leaving,” says Nathaniel Manning, co-founder of Kettle, an insurance startup that uses a type of artificial intelligence to understand how climate change affects risk, and to sell property insurance accordingly. “As a startup in this space, this is great for us.”
Kettle uses its algorithms to evaluate every property in the state of California, and line them up from lowest to highest risk of destruction by wildfire. The company has been running its models since 2020, and the results are telling. “In the last three fire seasons, about 20,500 properties have burned down,” says Manning. “Of those, 98% are in the top 25% riskiest parts of the state, according to our model.”
Kettle offers insurance for commercial properties, excess insurance for high-value homes, and reinsurance coverage. It currently covers about $200 million in insured value. Kettle’s coverage is often the least-expensive option for customers who can’t get coverage from any other carrier, and would otherwise have to rely on California’s state insurance of last resort, known as the Fair Plan, says Manning.
The internet of disasters
Another way to insure properties in a world where insurance companies can no longer afford to cover the full replacement cost of a building is parametric insurance. The customer and the insurance company agree on a flat payout if a specific event takes place.
In the case of the Postlane warehouse, the coverage depends on a sensor that has its own power source and connects to a cell tower over a low-power, long-range wireless standard known as LTE-M, says FloodFlash Chief Executive Adam Rimmer.
When it detects at least 8 inches of flooding, the payment from FloodFlash to Postlane is made more or less immediately. There’s no need for the time-consuming process of estimating how much damage has occurred, known as adjustment, says Matthew Raymer, Postlane’s chief risk officer.
This approach limits the liability of the insurer, and in turn can lead to lower premiums for the insured. Whether or not the agreed-upon sum is enough to cover the actual damage is up to the customer to determine—but that’s something they should be doing anyway, even with conventional insurance, says Raymer.
Other parametric insurance startups, like Plover Parametrics, rely on data from third parties to determine whether a particular event occurred, including wind, precipitation, temperature, fire, earthquakes and tornadoes.
“We are peril-agnostic,” says Clarke, head of the company. “What we have built is basically software that makes it easier to offer any kind of product for any kind of peril.” Plover doesn’t actually use its own capital to insure customers. Rather, it uses data and software to structure deals that other insurers might want to take on, but wouldn’t otherwise normally have the resources to do so.
All of these tools allow some insurers to continue to offer coverage even as the climate crisis worsens and the damage done by extreme weather increases. But they don’t change the fact that customers will have to pay more for the same or less coverage—or else figure out how to insure properties in new ways that might not fully cover every eventuality.
And no insurance will change the fact that some properties will have to be abandoned as sea levels rise and storms intensify. “I don’t think insurance is anything but a Band-Aid frankly in any context,” says Clarke.