Inside the insurance industry’s destruction lab

This is Planet Money from NPR. One muggy hot morning in rural South Carolina a few months ago, I found myself walking into an enormous warehouse in the middle of the countryside. It was there to meet a woman named Anne Cope, who was on a very particular mission. Do you ever get nervous before setting a house on fire? Well, no, I don't get nervous before setting a house on fire, because once you light it, you're going to get whatever you're going to get. Anne tells me that over the last decade or so, she and her co-conspirators have torched dozens of household structures, sheds, decks, all the way up to full on single-family homes. And while that may sound like the work of some kind of serial arsonist, Anne explains that she and her colleagues specifically build these structures in order to destroy them. Absolutely. We want to sacrifice the structures here in the lab. And we outfit them sofas, magazines, potato chips in the kitchen cabinets. We outfit them exactly the way they would be in a normal community, a normal environment, because we need to know when that stuff burns, what kind of heat, what kind of embers, what kind of threat is that creating for the rest of the community so that we can burn fewer of our neighborhoods. You can think of this place kind of like a destruction sandbox. Here structural engineers like Anne Cope regularly subject homes to trials not only by fire, but also by hail and hurricane force wind. They want to see how different building techniques stand up to stress in the time of catastrophic climate change. So you really have like a front row view of Mother Nature's worst. Like a machine for creating acts of God. It is. Now the act of God we are here to witness today is a little bit less catastrophic than a full blown hurricane, but it's still pretty biblical. Anne and her colleagues are preparing to burn a bush. Right now we're working on wildfire and so we're looking at what is the impact of having beautiful landscaping right next to your house. Like would a well-watered fight guess five feet from your house make it more or less likely to burn during a wildfire. What about a dehydrated juniper? I want to see which kinds of plants at which distances and drought levels might pose a threat to the homeowners who plant them. The experiment is going to take place in what I like to call the main destruction lab, which is basically a massive concrete airplane hangar of a space. The ceiling is six stories tall. On one side there's a looming wall of industrial fans. There's 105 of them. Each one is a 350 horsepower motor. It's a lot of horses. It is. A lot of horses, so that's how we can recreate a category through a hurricane right here. And on the other end of the hangar you can see what looks like a tiny home. This is what they're going to use to test how much danger each burning plant poses. The front of the house is covered in heat sensors and about five feet in front of it there's a single green bush that has been sentenced to burn. Off in the distance beyond the house I spy with my little eye a couple rows of other plants waiting for their turn in the hot seat. Are those the test subjects? Yes, those are all the plants and they are coming to meet their demise eventually. They're meeting their maker? Yes, yes, dead plant walking. We retreat to a safe distance before two men in firefighting suits carrying what look like tiny flame throwers walk over to the mulch pile. Oh, is it? It's getting close. I'm really rooting for the plant. I know that's not totally scientifically sound. The men aim their little flame throwers at the ground and set the mulch ablaze. They're lighting it on fire right now. Yes. As to the question of who would go to the trouble and expense of doing all this, of planting and burning dozens of plants and building and destroying a small pounds worth of homes and sheds and decks, this destruction lab is not administered by FEMA or any other part of the federal government. It's not a part of South Carolina's firefighting infrastructure or a wing of the construction industry. So this playground of catastrophic mayhem is funded by some of the biggest insurance companies in the world. Hello and welcome to Planet Money. I'm Alexi Horowitz-Gazzy. We're living in a time when climate change has made billion dollar disasters into basically a monthly occurrence. And that unprecedented level of destruction and loss has pushed insurers toward a sort of existential crisis. Today on the show, why the insurance industry decided to create crash test dummies for houses, how that research is changing the way we think about building new houses and fortifying old ones against disaster, and what all of this reveals about the ways the insurance industry has been quietly shaping the world around us for over a century. Before I started reporting about how climate change is changing the insurance market, I thought about insurance the way a lot of people do, as little as humanly possible. Insurance was something I needed to get a mortgage or drive a car, but then basically put out of my mind. That is, until I spoke to one of the people responsible for building that home destruction lab in South Carolina, a woman named Julie Rochman. So I'm an insurance geek. I always tell people I'm bilingual, I speak insurance and I speak English. Julie was actually raised in an insurance speaking household, where dad was an insurance broker. And after grad school, she moved to DC and ended up falling into the family business. Julie said she was drawn in particular to the property casualty part of the insurance industry, which covers things like commercial and residential property and car insurance. Property casualty insurance is the oxygen of commerce, right? We live in an entrepreneurial economy. The whole basis of an entrepreneurial economy is risk taking. And you have to lay off or spread risk. No one person can handle all of the financial and physical risks involved in most businesses. The whole modern global economy couldn't even gotten started without the idea of pooling risk. Julie explains that is when early corporations like the East India Company or Hudson's Bay Company started undertaking these expensive, super risky colonial expeditions across the sea. Investors would front the money to procure ships and fill them with all sorts of goods to trade. But those ships in their cargo went missing or sank pretty often. So a market developed for other investors to assume some of that risk by placing a sort of bet. They would offer to pay a certain amount of money if the ship did sink. And in exchange, they would get a smaller chunk of money if the ship made it. When a ship was slated to set sail, the owners would write up the details on a piece of parchment. And the people who wanted to take on that risk would write their names and the amount of risk they were assuming under the ship's name. This is reportedly where the term underwriters comes from. And all of this business, Julie explains, would be transacted at these bustling coffee houses near the ports, like a place called Lloyd's by the Thames. That's what Lloyd's London was. It was a coffee house. And so coffee houses where rich people gathered became marketplaces where somebody would say, oh, your ship is leaving tomorrow. What's on it? Oh, it's gold. It's going to the new world to pay for the tobacco and sugar that's coming back. Oh, OK, some people would ensure the gold on the way over. They would share that risk. And then other people might step up, could have been some of the same people and say, OK, I'll take part of the risk that the products that are coming back make it here safely. And so it was just a way to make sure that everybody pun intended state of float. And the US Julie explains, the modern insurance industry developed around a different kind of risk. The insurance industry in this country was really born out of fire. The houses in the colonial era were mostly built out of wood. People used fire to cook and light and heat their homes. And it was, as you can imagine, a highly flammable combination. And one of the first American insurance companies was co-founded by none other than Benjamin Franklin. It was aptly, if a bit windally named, the Philadelphia contribution ship for the insurance of houses from loss by fire. And the basic idea then, as now, was essentially to pool as much money as possible in order to shield any one person from being financially ruined by a big fire, like a mutual aid agreement. And that was the principle at play as new insurance companies started to pop up around the country, organized around specific regions or communities looking to hedge against the desistitudes of fate. Most insurance companies were started as mutual companies owned by their policyholders. So it was a county, mutual in Indiana or Lutheran mutual came through a church. I mean, they were all these specialty companies and then stayed farm nationwide. I mean, farmers, these were all started by farmers. Julie says these insurance companies, which had started as this form of mutual aid, got bigger and bigger as they signed up more and more people paying into a common pool of money. And then it is that pool of money that the insurers manage. They invest it. They hope that they can grow it because then they will pay outlosses with that money. As more and more Americans moved west or into coastal towns and places like the Gulf Coast, so too did the financial liabilities and profit opportunities of the insurance industry. And so insurers started creating new kinds of policies to cover that risk. Things like tornado insurance in the late 19th century or earthquake insurance in California in the early 20th. And in this way, Julie says you can think of the property casualty insurance industry almost like this giant financial mirror, reflecting and responding to whatever forms of peril we all collectively face at any given time. And that peril doesn't just come from natural disasters, it also comes from human innovation. Once the industrial relation hit and there were more machines. Those machines need to work and sometimes they injure people and so workers come comes along and it is just literally like every time There's a new Industrial sector or a new product or a new service We all go back to Lloyd's of London which is we have to share the risk right we can't allow a Person or a company to be destroyed because of an accident and it was one new technology in particular that inspired the insurance industry to radically Rethink its approach to risk a new approach that would eventually lead to the destruction lab. I visited in South Carolina Julie says it was to ward the tail end of the industrial revolution when the insurance industry started Banding together to do its own experimental research to figure out not only how to better price risk But to find ways to actually reduce it out in the real world that new technology Julie explains was Electricity when electricity went into homes during the Victorian era the wires were not all that safe They were cloth wrapped, you know the switches the bulbs everything could explode or overheat and cause fires And so the industry the underwriters the people who bear risk it got together and created a laboratory in which they could do Testing rigorous objective testing of these products this insurance funded laboratory in Chicago was called Underwriters laboratories or UL for short it started by testing out these new electrical wiring systems to figure out Which were more dangerous than others? But they also eventually expanded their mission to include many of the newfangled appliances and products that started to pour onto the market They wanted to determine which might be likely to electrocute somebody or start a house fire and if you passed the testing Then UL would put their mark on your product the UL product mark meant essentially that somebody who is looking out for your safety Says this particular thing is safe now insurance companies did not have the power to Dictate new safety regulations, but they did have the power to deny coverage to buildings or homes that did not choose UL certified wiring and they could convince mega retailers like sears to use UL safety mark to decide in part Which electronics and appliances to stock and in fact to this day if you look at the electronics around your house Odds are high that hidden somewhere near the serial number You'll still be able to find this little circle with the letters UL inside of it for underwriters laboratories Meaning this private industry group has been able to exert influence over infrastructural and consumer standards You might usually associate with the government So that was one of the insurance industry's first major forays into experimental research and intervention in the market Julie says the next big example was automobiles As cars took over the roads, insurance companies were on the hook for more and more accidents And so they eventually funded their own crash test facilities And the research from those facilities and videos of crash test dummies flying through windshields They helped push car companies to change the design of things like headrests and airbags But Julie says the idea that the insurance industry might apply this experimental approach to the buildings we all live and work in Started percolating about a decade later in the mid-2000s Around then she started hearing rumblings about the latest and perhaps biggest new form of risk that the insurance industry now had to contend with The threat of increasingly frequent and destructive natural disasters We all went through Katrina Hurricane Katrina and saw the tremendous loss of life Almost the death of a city And then in 2007 there was the Witch Creek wildfire in San Diego County that was a billion dollar wildfire And so people in the industry could see that these were becoming mega events Anything over a billion dollars we typically would call mega event And the industry was looking at losses that were being driven in new ways and to new levels And that's not sustainable Like if catastrophes keep going like this there won't be a property casualty insurance industry anymore So Julie says a non-profit insurance trade association called the Institute for Business and Housing Safety or IBHS Decided they wanted to help engineer a solution to this problem They decided it was time to figure out how to do for buildings what the industry had already done for cars and electronics To create their own destruction lab where they could subject various kinds of structures to various kinds of disaster In order to learn what building techniques and materials might ultimately reduce all this catastrophic loss The dream of IBHS was to create a rigorously meticulously sourced set of data points That would inform builders, architects, developers, public policy makers, the media and obviously consumers So that they could make better choices about where and how they live and work So that losses would not occur on that same catastrophic level as often as they were starting to occur And the industry looked around for someone who felt like being a part of that and leading the charge and it ended up being me Julie had experience doing just this kind of work. She'd worked with the insurance industry's car crash research group for several years So becoming the CEO of IBHS was a natural step But on Julie's first day on the job the board told her that it was now her responsibility to raise the full budget to get the facility up and running She'd have to raise a total of $40 million dollars And I'd never done fundraising And I just thought what the heck have I done Julie starts reaching out to insurance companies around the country to start raising money And she knew they all had a vested interest in researching the science of property destruction But each of them also had a different risk portfolio So depending on where they operated and what kind of disasters they were trying to cover Julie had to cater her pitch accordingly So if I was talking to an insurance company in Florida, we talked about hurricanes If I was talking to an insurance company in Colorado, I was talking about high winds and hail If I was talking to an insurance company in Nebraska or Iowa or Illinois, I was talking about tornadoes, high winds and hail winter weather And those are very different conversations to have Julie explained to each company that if they invested in IBHS's research they would eventually see returns in a couple ways First, their research would help identify risky materials and building techniques that posed a financial risk to the insurers This new information would give them a more accurate way to look at which houses they offered coverage to and at what rate So if someone had a roof that was now proven to be more likely to cost the insurers way more in the event of a storm They might choose not to cover that property and they could offer discounts for people whose construction decisions made them a lower risk And second, insurers have been paying to rebuild people's houses in the same way in the same places Again and again after successive disasters There just wasn't enough science to know why some houses survived a disaster and some didn't If IBHS could sway building practices and market dynamics toward products that were less likely to fail The insurers would ultimately have to pay less to make their customers whole because there would be less property damage overall Julie says she managed to get funding from a few of the biggest insurance companies relatively early on But others needed a bit more convincing I would go out and do presentations and I would list all the funders and originally it was one slide, right? And then it got longer and then I would say you may notice some names missing on these slides You know when I would go to a company and say you don't see your name up there, right? And shame definitely does work in a competitive environment Pure pressure is a hell of a drug It is Julie says getting the facility built took a couple of years of intensive lobbying At the same time her counterpart at IBHS, the chief engineer, was busy figuring out things like How do you actually create a tiny artificial hurricane inside a building? Or how do you measure how many Embers it takes to light a roof on fire? Finally, in the fall of 2010, the lab was ready for destruction Julie and her colleagues invited a group of their investors, insurance company executives, to come see the inaugural set of test demonstrations They set up a pair of two story houses in the test chamber One of them was built to the standard code and one was built with extra fortifications They then programmed the giant wall of fans to reproduce the wind patterns of a hurricane And when they turned it on, something unexpected happened to the unfortified house You can actually still see this in a video they took at the time The whole house blew out the back door of the test chamber It was literally in a pile, you know like in a field in the pile Julie and the IBHS scientists hadn't expected the house to completely blow away But they still were able to learn how small differences in construction could lead to vastly different outcomes The house that survived, it had been built with fortifications that cost less than 5% more than the house that ended up a pile of debris The insurance executives were impressed And over the next several years, Julie and company built out their program according to the specific research requests from the various insurance companies backing IBHS The so-called member companies. In exchange for dues, sometimes in the millions of dollars, member companies get exclusive access to the full range of IBHS's research Though some portion is also released to the public And when you think about all this from an insurance market perspective, there's something sort of ongoing on here Like usually all these companies are locked in a heated competition for customers That's why we're all bombarded by ads with talking geckos or annoying jingles Yet in the case of this IBHS destruction lab, here they all are, congenially cooperating in order to fund research which could benefit them all Which means these companies are also constantly walking a sort of legal tightrope Because under U.S. antitrust law, which is meant to promote competition, competitors are strictly forbidden from coordinating on things like their pricing strategies or market segmentation Because of the risk of collusion How does the specter of anti-trust law hang over this. - Anti-trust law is at the center of all things property-casually insurance. And all meetings start with what I call the anti-trust invocation where before anybody says anything. There's a lawyer or somebody acting as the anti-trust counsel from outside. They will specifically say at the start of every meeting. We are forbidden to talk about company operations and business practices. If anybody gets anywhere close to the line, we will stop the conversation immediately. And fortunately, I haven't seen a lot of people stopping those conversations over 40 years, but occasionally, you will see a whole bunch of people stand up around and table and go, (laughs) no, no, the floor is lava. - Yes, the floor is lava. - No, it's true that the research IBS does provides all these companies with a common body of knowledge they can use to tweak their business models. But each of them is also creating super secret formulas from their own specialized data sets to try to determine whether to offer coverage and how much to price their premiums. So they can get an edge on their competitors. Julie says, "IBHS figured out a way for their funders to communicate their needs without breaking the anti-trust rules, and IBHS has been able to build out a dizzying array of experimental programs, some of which they've published for the general public." - It's to show people the data and to teach people which questions to ask so that overall, we are putting in place better building stock. We're not just putting the same types of buildings in the same vulnerable places where they will be demolished by the same natural forces again and again and again. And that's what we had been doing. And we've basically been building with steel and brick and what, you know, in the same way for a couple hundred years. - Julie retired from IBHS back in 2018. The Research Center has continued to build a growing body of research that insurance companies can in theory use to evaluate their premium pricing or to choose who to approve or deny for coverage. It's also research that developers can use to decide what kinds of materials they use and that governments might use to update their building codes. So this one corner of the private sector, well, this nonprofit research organization for the insurance industry is producing a sort of semi-public good. It's only semi because not all of their research goes to the public, it is a membership organization after all. But the information they do share can benefit a bunch of homeowners and neighborhoods in cities by potentially reducing the amount of destruction from storms and wildfires and other mega events, which also happens to help the insurance company's bottom line. After the break, we head back to South Carolina to see how these researchers put the destruction in the destruction lab. We hear some of the lessons they've learned, how their research is being put to use on the ground and how the insurance market itself might be making it harder to convince people to fortify their homes against disaster. When I visited the insurance industry's destruction lab in South Carolina a few months ago, I got to see all the catastrophic research programs in action and to find out what they've learned from all this high octane experimentation. IBHS chief engineer Ann Cope took me on a little tour of destruction. There was of course the giant hanger for lighting houses on fire or blasting them with wind and rain, but outside the facility there were even stranger sites. Endpoints to a curious cluster of disembodied roofs standing on stilts in the distance. It's something they call the roof aging farm. Those are asphalt shingle roofs just sitting out here in the nice South Carolina sun and we're watching what's happening to those shingles over time. Ann says the plan is to leave them out there in the elements for as long as 20 years. - You're just waiting for them to ripen? - No, well, that's the kind of stuff that we're studying there. - Periodically, some samples from those roofs are also harvested and yes, that is the technical term and then brought into a different part of the research campus. - Welcome into the Hale Lab. This is where we do all of our Hale Impact testing on building materials. - Hale Scientist Jake Sorber explains that parts of those roofs are then mounted on a wall and turned into targets for artificially manufactured a Hale Stones of varying density and speed. - Hale is actually being made right now. It looks like he's about ready to be done with this specific rod. - These lab-grown Hale Stones are then shot out of a cannon. - Ready, shoot, three, two, one. - Wow. - The Hale team then examines the piece of roof they just pelted using a 3D scanner. And Jake says this kind of research addresses a growing problem for the insurance industry. See, before 2008, Hale was a relatively small line item on insurance loss profiles, often averaging below a billion dollars a year. But in 2008, Hale losses started to climb dramatically. These days, Hale damage can cost insurers tens of billions of dollars. And the choice of shingles that somebody uses on their house could mean a difference of tens of thousands of dollars in damages. So now, in part because of this kind of Hale Lab research, insurers can evaluate their coverage decisions according to which shingle somebody has. And builders and homeowners have more information about whether bansier shingles might be worth it for lower insurance and repair costs. Chief engineer Ann Cope tells me the destruction lab has discovered all sorts of news you can use. For instance, they've tested out different building techniques for how much wind-driven rain might leak into your attic during a hurricane and found that adding an extra layer of relatively cheap waterproofing to your roof, sealing the deck, that could prevent up to 95% of that kind of damage. Ann says these techniques have already made a big difference in communities where they've been implemented at scale. Take the state of Alabama. Insurance regulators there offered direct grants to homeowners to fortify their roofs to an IBHS standard to prevent wind and rain from getting in, but mandated insurance company discounts for people who installed them. And over 50,000 people in Alabama have fortified their roofs. Ann says it's already helped homeowners there. - Hurricane Sally comes across right in the same area where all of this fortified development had occurred. And the IBHS fortified roofs were the ones that were in great shape. People were still in their homes. Neighbors who did not have a fortified roof had tarps, had damage, had water, and the fortified roofs were the ones that kept the community together. It's a triple win. - A triple win, Ann says, because a roof of less damage means less disruption for an individual homeowner. More people are able to stay in the community after a major storm and less damages for the insurance companies to pay out. As for what homeowners can do to defend their homes against wildfire, that gets us back to the experiment I'd come to the destruction lab to witness. What I like to call Exodus 3-4, or the experiment of the burning bushes. Sharini Munroy is the wildfire researcher charged with executing that experiment. - We have been doing this many times, maybe 50 so far. We have another 50 to go, at least. How many bushes have to die before you're satisfied? - Oh, it's what we have to pay to understand how these things behave, right? - It's through research like this that the industry, along with communities around the country, have settled on the idea that one of the single most important ways of hardening a home against the threat of wildfire is by limiting the flammable material within five feet of the house, creating a sort of defensive perimeter. And this particular defensive technique is now even being mandated for millions of people. Individual towns and counties, even the state of California, have adopted this idea into their building codes. And some insurance companies have made it a requirement in order to qualify for coverage. IBHS has a whole list of other recommendations for homeowners living in places with a high risk for wildfires. Homeowners can limit embers sneaking into their attic and lighting the house on fire by replacing their vents or adding much smaller metal mesh. They've also found that wooden fences can act sort of like a wick in a wildfire, spreading the blaze from home to home in a neighborhood. So swapping wood fences out for metal ones can make it less likely the fire will spread. Unfortunately, the scientific consensus here is that in order to significantly reduce a home's risk of burning down, you really should do all of this stuff, which is something a lot of people simply can't afford to do. Now, IBHS is trying to incentivize homeowners to do everything by offering a certification called wildfire prepared. When a homeowner gets that certification, it should, in theory, signal to insurers how much work they've put into mitigating potential losses. And ideally, it might translate to lower premiums. But I've been reporting on the world of wildfire and insurance for a few months now, and talked a lot of homeowners dealing with this issue in California, which is facing some of the most extreme and expensive wildfire risk. And I found that some communities that are trying to follow IBHS's new building standards to the letter of the law are still struggling with skyrocketing insurance premiums. Some homeowners have seen price increases of almost 1,000%. Or they've gotten dropped from their insurance altogether. I ask IBHS's Ann Cope about this. How do we make sense of that problem of like that somebody could be doing everything they possibly could to try to use IBHS science to kind of recreate their homes and still not be finding a reasonable place in the insurance market? And so, there are multiple pieces to all of that. Individual properties one by one are part of the complicated macro scale. And says it's important to understand the physical risks of any given property are only one part of the multi-dimensional puzzle that goes into silence. someone's individual insurability. There's also the question of what the neighbors have or haven't done to mitigate the risk of wildfire. Stopping the spread of fire is a collective action problem, because if the neighbors don't harden their homes, those homes can still pose a big threat to yours. And there are market issues, like whether a given insurer is overexposed because it's covering too many homes in one geographic area at risk of burning down all at once. And says IBHS can't solve all those issues. Others are going to be able to talk about the economic impacts and the geographic impacts. And we're just going to keep bringing the best possible credible science to say, to make the resilience of this structure high, here's what you need to do. And says it's IBHS's job to help build the scientific research for everybody else to act on. And the fact that catastrophic mega-fire might be breaking the insurance market itself, that is a problem that goes way beyond the destruction lab. It's a problem playing out in forests and communities all across the country. [MUSIC PLAYING] Next week on the show, Planet Money goes to perhaps the most dire conscious town in America to see how the insurance market has broken down and how the people there are trying to build it back up. [MUSIC PLAYING] If you want more on how economics shapes the world around you, sign up for the Planet Money newsletter. Sometimes it's an original essay or an interview or a curated digest of recent econ papers we liked. Our newsletter writer, Greg Wazalski, puts his heart and his giant brain into it every week. We see you, Greg. Sign up at npr.org/PlanetMoneyNewsletter or click the link in the show notes. And if you already subscribed and liked it, hit Reply and let Greg know what you think. [MUSIC PLAYING] This episode was produced by Vido Emanuel and edited by Jess Jeng. It was engineered by Anli Huang and Travis Hagan. And fact-checked by Sierra Juarez with help from Emma Peasley. Alex Goldmark is our executive producer. I'm Alexi Horowitz-Gazi. This is NPR. Thanks for listening. [MUSIC PLAYING] [BLANK_AUDIO]