Welcome, everyone. Today, we're discussing a critical intersection, and that's climate risk and housing affordability. What we're really exploring is with extreme weather events becoming more frequent and severe, how is this reshaping the mortgage market and homeowner decisions? To answer that question, I'm joined today by Larry Lawrence from Ice Climate and Andy Walden from the Ice Mortgage Data side. Welcome to you guys. Appreciate it. Thanks for having us. Glad to be here. So glad to have you both on. Before we dive into the data part of this, I'd love to hear a little bit about what your teams do and what motivates your work. So Larry, let's start with you. What is Ice Climate and what gets you excited about this space? All right. Lucky me. So I'm Larry Lawrence. Ice Climate is the climate and geospatial intelligence division of ice. We're focused on bringing the same kind of transparency to climate risk analytics or extreme weather analytics, if you want to call it that, that ice has brought to other parts of the financial markets over the last few decades. We're a team of data scientists, climate and financial modelers focused on building large data sets of location-specific information, thinking about where buildings are located, where offices are located, where data centers are located, a topical thing. And then we link those to the financial assets that our clients own and help them think about how these different types of risks, whether it's extreme weather today, happening now or in the future, could potentially have an impact on the revenue of these different assets. Our mission is really to help lenders, insurers, homers make better decisions by understanding these risks, these exposures at the property level, the loan level, and at the asset level for the balance sheets that they manage. So in particular, we think this is pretty relevant today because most people think about climate risk and think this is a future event, a future thing to pay attention to. But what we found is that it's impacting different parts of the housing market today, and it's something that investors are paying attention to. Larry, I think that's really important, too, because I also think there's the sort of idea of like, and we can't know what's coming. So, like, let's just deal with what we can do. But what you do there with the data is you are surfacing the things that can be done now and how it might be impacting right now. So super interesting. Andy, what about you? And as you know, sir, I head up the mortgage and housing market research team at ICE. And so I get the kind of distinct pleasure of getting to play with the data sets across the organization. And in many cases, a lot of traditional data sets that we typically think about mortgage origination data, mortgage performance, housing market trends and public records and all of those things that we're pretty experienced at playing with in the industry. But obviously, one that's coming to the forefront is obviously connecting that traditional data with what Larry's team is doing over on the climate side, looking at how climate's impacting mortgage performance and mortgage related risk in the market, and then tying property insurance data into that as well. And really the intersection between climate risk, property insurance coverage and overall default risk in the market. And so I get to discuss that a lot internally here at ICE and get kind of the pleasure of discussing it externally in forums like this as well. So excited to do it. I love that. Well, let's dive into some of that research, right? So what can ICE's data tell us about the impacts of extreme weather events and climate risk on mortgage performance? And specifically, Larry, your team's recent white paper examined delinquency rates after natural hazard and extreme weather events, which I think is really interesting. So let's start with that. Yeah, no, it is. It is really interesting. I think it sort of really points to the fact that this is impacting everything now. And it's a today problem, not a tomorrow problem. So our research team analyzed mortgage delinquency patterns following hundreds of sort of extreme weather events over the last 14 years. These events include, think of wildfires, hurricanes, tropical storms, even floods. And we did that analysis since looking from 2012 to current. On average, what we see is that after a wildfire, hurricane, or flood event, short-term mortgage delinquency probability increased substantially in the following one to three months. So that short-term shock is pretty extreme. Over longer timescales, loans in higher risk areas, think of loans that have sort of higher elevated risk from flood, those experience pretty elevated probabilities of severe delinquency over the longer term, even after controlling for different borrower characteristics. So these climate-related impacts on the housing market, again, it's happening now. They're creating sort of short-term stress for servicers, local banks, even credit unions, as well as creating more concerns for those longer-term investors who are concerned about the performance of whole loans or mortgage insurers and others like that, that sort of try to insure for longer term. So we're seeing pretty significant shocks following these extreme weather events in the short term, especially. Andy, that's one side of the coin. What do you see when you look at these same things? Yeah. I mean, if we look at it from the property insurance side, you see a very similar storyline playing out. I think we've all heard kind of the headline numbers for property insurance. It's up 70% over the last six years. It's the single fastest growing subcomponent of a mortgage payment. And it's not just a Florida or Louisiana or Texas story anymore. I was at a conference in Nebraska a couple of weeks ago. We were talking about how Nebraska has the fourth most expensive property insurance per dollar of coverage anywhere in the country. Oklahoma is right there as well. So you're seeing it up the Midwestern region of the country. California, after the wildfires, you're seeing the fastest growth in property insurance costs anywhere in the country now out in California. And so really, regardless of where you sit from a geographical perspective, starting to see some of that exposure coming into the market. And we did a project with Larry's team, what was that Larry, two months ago or so, looking at property insurance costs specifically across the country. And we looked at not just the traditional PITI that we all think about, but they rolled in water costs and energy costs and everything that we're paying to own a home over time. And what they found was that the larger share of that expense that goes towards property insurance, the more likely you are to be behind on your mortgage payment. And you can see it across credit score bands. If you look at low credit score versus mid to high credit score bars, you see it similarly across those groups You can see it across DTI bands When you look at it across various geographies you seeing that same uptick in correlation where the more you spending on property insurance the more likely you are to be behind on mortgage payments And so you're seeing it not only on the climate side, but you can use that property insurance data as another risk attribute almost in terms of overall credit risk in the market as well. I like that you broadened it out from maybe the coastal areas where everyone thinks like, oh, okay, well, that's normal. Not normal, but like we expected hurricanes, flooding. I was just in Colorado for about 10 days and it has never been where I was around the Denver area. I have never seen it so brown at this time of year. And I think they just didn't get a lot of snow. They haven't had a lot of rain. And you know what's coming, right? That's a huge fire risk. In some ways, if you have that kind of hotter temperatures, that's a risk for more storms. It's crazy how it affects everywhere. Yeah, it's true. And I mean, I live here in Denver and I'll give you a little inside story from my own life. I got my escrow restatement last week. My property insurance jumped by $900 just this year. I've had zero claims. I've got a co-worker that lives 10 miles from here, kind of more on the Eastern Plains. It was just dropped from their insurance provider. And so you're right, it is everywhere. And we're feeling it here in Colorado as much as anybody else. So let's talk about how homeowners are looking at this, how they're considering this when they're thinking about where to buy, what to buy. So beyond the credit stress specifically, how are homeowners considering this as they make their decisions? Yeah, let me give you maybe a couple of different examples that I've seen recently. One came out of our 2026 Borrower Insight Survey. So every year at ICE, we reach out to a bunch of recent mortgage takers, a bunch of recent homebuyers, as well as renters that could be prospective homebuyers in the near term here and ask them about what they're thinking about the market, what's important to them, what was important to them throughout the homebuying process, etc. I asked them a question about climate risk this year and how important is climate when you're making a home buying decision. And three quarters of homeowners said that it was impactful and part of that decision making process. 40% said it was very or extremely important to them. And it was pretty consistent across geography. So the point you were just making pretty consistent across geographies as well. It wasn't just the south. It wasn't just the west. It was saying that climate's important. And so what that tells me is we need to include more of that climate data in places that are actionable throughout that loan taking process, whether it's upfront when you're buying the home, whether it's through the loan origination process or servicing process, the more actionable information that we can put out there in the market, the better. The other piece that I saw, very clear decisioning processes being made by homeowners and mortgage holders was around insurance switching in recent years as well. If you look back over the last 10 to 15 years in the market, mid-20 teens, about 7.5% of mortgage holders switched insurance providers every year. It was a pretty consistent churn. We look at what's happened over the last few years as insurance has become more expensive, seeing more and more folks shop for and switch insurance providers. It was all the way up to 11.5% last year. We saw a record number of folks switching insurance. And it's happening for a couple of different reasons. One, maybe they're being dropped, right? We're seeing a lot of those cases across the country, Florida, California. Again, I mentioned here in Colorado, we're seeing some of that as well. So you're either being dropped and you're forced to go get new insurance, or you're seeing enough of an increase in your insurance on a year-over-year basis that you're going out there and shopping, you're shifting your coverage, you're shifting your deductibles, you're finding more affordable and more attractive insurance options out there in the market. And so you're seeing more of that activity than we've seen historically as well, which again, you go back to how do we help homeowners in that journey? It's how do we help them shop for insurance upfront through that loan origination process? So we've built kind of APIs into the lending process to help folks shop and obtain the most optimal insurance for them. And then even throughout the life of the loan, embedding that in your servicing platforms and other technologies so that as folks go, for folks like me that just had my insurance raised after being in my mortgage for five years, going out there and providing them a frictionless way to shop and improve those policies and reduce that expense as well. It's such a shift because I remember, I mean, even say six, seven years ago, you could see the data, you can see what's happening. And yet it did not feel like it had gotten to maybe your average home buyer or homeowner. But that's because, you know, I mean, a lot has happened in the last six, seven years. And I do think that, as you said, I mean, people are aware of it because their insurance is changing, because their neighbor's insurance is changing. Like, it's not something that's just like, oh, well, that's happening somewhere to someone. It's like, no, it's happening to me. It's everyday conversation. I've had a dozen of them over the last two months outside of our industry, just talking to neighbors and family members in different parts of the country. It's everywhere and folks are shifting their behavior because of it. Larry, what are you seeing? Yeah, I mean, I can speak to sort of the home value side of the story because our recent research piece dug into this quite a bit. So what our research suggests is that properties in sort of flood prone areas are experiencing a, let's say, a 0.2 to 0.4% slower annual home price appreciation on average compared to like similar homes and areas without flood exposure. So flood risk is quietly shaven, you know, 0.2 to 0.4% per year of home price appreciation in high risk zip codes. Now that may not seem like a lot, but let's dig into what that means. So on average for a home that is valued at, you know, $250,000 in a high risk flood zone in say 2013, maybe worth $15,000 less today. And then it's sort of zero risk equivalent. So a home in like a, you know, zero flood risk area. In total, in the broader market, that's a $31 billion loss in residential value since 2013. So if you look at the trajectory, the decade of strong appreciation, the discount on these home price appreciation has kind of been masked a little bit. But the risk is what comes next is especially important. Investors in mortgage credit, in RMBS who may have priced this in, or will need to price that in, will need more data to help them do that. So, you know, I know I mentioned 0.2 to 0.4% that can get lost, but that's $31 billion in home price value that's quietly being shaved off in the market because of exposure to high flood risk zone No that super important You know one of the things I love about this conversation we having is that like I said before like climate risk can just feel like so nebulous So what is ICE doing to make climate data actionable It really means something to lenders, to borrowers. And really our goal there is to make it available in the form and fashion that's the most useful for our business partners. And in some cases, that's through traditional data deliveries, those flat files of climate risk data or property insurance data or mortgage performance data in the forms that you would be used to injecting those and allowing folks to pull those into their traditional workflows. In some cases, it's embedding that data into our tech stack here in ICE, whether it's through an origination platform or through our servicing platform as well. So really, it's putting that data at your fingertips in the way that you want to get it. And there's a number of different ways to do that. When we look at the industry and how they're utilizing this data, I tend to think of it in kind of a before, during and after an event type of structure. So before an event takes place, before the hurricane even hits, before the wildfire even starts, it's really a lot of portfolio and risk management. So running your, if you're a lender, looking at your lending footprint and where you typically lend in the market and what the climate risks are there, and looking at climate risks on specific properties that you may be lending on. If you're a servicer or a capital markets participant, it's looking at your servicing portfolio, your whole loan or MBS portfolio, and looking at your climate risk. Are you overweight climate risk? Are you underweight climate risk? How do you stack up to the market at large? And in a lot of cases, they're using that as part of the due diligence process as they're acquiring new MSRs, buying new whole loans. We all think of due diligence as looking at the credit quality and everything associated with those loans. You're seeing more and more folks, you're almost on the outside looking in if you're not utilizing climate data as part of that due diligence process to make sure that folks aren't offloading their climate risk out there into the market. And you're unsuspectingly picking up a big chunk of climate risk, either in your servicing portfolio or your whole loan portfolio as well. So that's a lot of what happens before these events hit. Once an event actually takes place and there's a wildfire or there's a hurricane, a lot of that is disaster alert. So Larry's team will feed you data overnight, whether it's at the parcel level or loan level. For a lender, it's how many loans do I have in flight that I'm halfway through that origination process I'm about to close on that may have just had the collateral impacted by a hurricane or a wildfire that I need to do an inspection before I close on that. On the servicing side, we used to rely heavily on FEMA county level data. Well, that's a pretty wide swath of your portfolio that you may have to weed through to figure out who is actually impacted. Having parcel level identification makes you a lot more efficient as a lender or as a servicer to provide borrower assistance and make sure that you're inspecting properties that need inspected and helping homeowners that need helping. And then Larry talked about some of the after event type of analytics, looking at what happens with home prices in the wake of these events, looking at public records data and permitting data in the wake of the wildfires. We've had a lot of questions around that and data polls around that and looking at how many folks are rebuilding, what are they rebuilding, how much insurance did they have, what kind of assistance programs would be best to help folks in the wake of this. And then using all of that data from post-event to then circle around to the front end and model that risk going forward. So the more of these events that take place, the more we know about them, the way that the market behaves, and we can more accurately model future risk as well. So a number of different ways that you can access the data. Everybody uses it a little bit differently depending on where you sit in the industry, but a lot of key benefits and a lot of ways to reduce risk exposure, regardless of where you are and create some efficiencies inside of your workflows as well. I think that parcel level information, it's so insane. And now it makes the FEMA, you know, like larger thing. It just looks like such a blunt instrument. It's like, you don't even know, you're like, okay, on this couple of blocks or something, it's like, we can get down to like, What does the fence look like? You know, what does your roof look like? What did the windows got blown out? I mean, it's a huge game changer. Yeah, we used to look at delinquencies that way, right? We would look at post-hurricane events and we would look at delinquencies inside of FEMA-declared counties. And now we can look at, you know, Larry was feeding us over data for the California wildfires. And we can look at inside the specific wildfire zone in the wildfire perimeter and look at delinquency differences just from the perimeter to 100 to 200 yards out from the perimeter. and you're seeing very distinct performance differences just in that distance. And then you see minimal impact outside of those zones. So you're right. It was a very blunt instrument before. It's a lot more precise now. And it makes us more accurate and more efficient as an industry as a whole. Andy covered a lot of ground. I would just say a couple of additional things. And I think you picked up on this quite well, is a lot of the challenges we've heard from clients is that they've previously relied on market level, zip code level averages to sort of get a sense of what risk is in certain areas. You know, and the data we have is really starting from the ground up. We are getting and mapping data to the parcel level, over 150 million parcels across the United States. So you're able to start at the asset level and build up from there because, you know, you look at one area, as you said, risk can change pretty drastically within a few blocks. So it's important to understand the nuances of that. So we've done a lot of work to map our data at the property level, at the loan level to help investors do that granular analysis and build it from the ground up, which is critically important. And in addition to that, whether or not you're interested in understanding, hey, as an event occurs, like the LA wildfire, I'll give you an example there. We had clients calling us almost every day during the two weeks that the LA wildfire occurred, because every single day went on the scope of damage increase, the surface area of impact increase, so clients were continually interested. Do I have any, how much of my assets are exposed host to this, I'd like to know now. So the capabilities we have can help them understand this in sort of near real time as events occur. Wildfire, it could be a flood, it could be a hurricane, or even if you're out of the US, a typhoon, we do that as well. So giving you that near real time analytics to help you manage risk in almost a real time basis is critically important. But beyond that, we can also help you think about your entire portfolio of assets and pinpoint where you may have outsized risk from these types of events to help direct how you may want to think about decisions in the future So I think it the granular data that critically important being able to help you with real events as they happen because that a lot more tangible for people and also helping you do portfolio level risk analytics so you understand where those hotspots in your portfolios are so you can use that to make decisions. You know, it's awful that we have more of these things happening, but I'm glad that it's happening at a time when we have the tech ability to get the information, understand what's happening and make decisions based on that. It's not just like, okay, all these random things are going to happen. And then it's just like a black box. No one knows, you know, at least we can delineate things and take action. So as we wrap up this conversation, I'd love to ask you, what do you think people get wrong about climate and the housing market? Like what's the biggest misconception out there, do you think? I mean, simply from my side, it's people think it's a future problem and people try to make it this thing that's sort of way down the road. But what the data shows, as I just described to you, these events are happening now. We had over $5 billion hurricanes in the last, not last year, but the previous year, hurricane season. So, the outsized impact these events are happening are significant. Someone is paying for that. The question is, is it you that's paying for that? You don't know without these analytics to understand which of your assets are exposed. So, the data shows it's already here. It's already impacting delinquencies. It's already impacting home value. And as Andy alluded to, it's already impacting home value decisions. So the question is whether or not you want to integrate this granular data to give you more information than not to help direct your decisions you make. I'll add a couple there. I mean, one we've already talked about, right? The misconception about geography and, you know, I think historically we've all thought about this as a coastal issue and it's a Florida, Louisiana problem and not everybody else. And obviously that's fairly clearly no longer the case. The other one is one that maybe I have, and I'll ask you guys if maybe you have this one as as well, but it's around flood insurance, which we haven't talked about a whole lot yet. And if we kind of rewind in time to kind of Larry and I kind of talking about impacts on the mortgage and housing market and what we wanted to look at originally with research projects, flood insurance and flood risk was at the bottom of the list for me. It was hurricanes and it was wildfires. And the more that we got into the data, the more eye-popping information came out on flood insurance. My thought up until a year ago was, if you have flood risk in your portfolio, you're in a flood zone, you have flood insurance. And yes, there's some risk there and it probably impacts demand on those particular properties and home price growth. But if you've got flood risk, you're covered by flood insurance. And that couldn't be farther from the truth when you start to get in and dig into this underlying data. And we started to match climate risk data for flood insurance with property insurance data with mortgage performance data. What we found was there were 5.3 million single family homes across the US that had one in 100 year flood risk. Sounds like that rarely ever happens. The older I get, the more I'm like, 100 years isn't that long. And the more you think of it in the context of a 30 year mortgage, you're like, well, that's one in four chance roughly of a property that you have in your portfolio being flooded at some point and sustaining some loss during the stated life of a mortgage. But then you look at how many of those 5.3 million have flood insurance. It's 15% of those loans that have one in 100 year flood risk actually are covered by flood insurance. Then you narrow it down and say, all right, what about just high risk homes? There are about 350,000 that have higher extreme flood risk. Only a third of those are covered by flood insurance. Two thirds of higher extreme flood risk properties aren't in a flood zone and don't carry flood risk. It doesn't make any sense and it doesn't match what my initial thought around flood insurance coverage in the US looks like. And then you start to look at why and you see that those initial flood zones that were created weren't really based on pluvial flooding. Pluvial flooding is a fancy way of saying rainfall and flash flood related flood risk. When you look at the climate data that Larry's team is using and the fact that they're using geospatial technology and topography maps, watershed models and things like that, you can look at what the true risk for flooding is. And it's very, very different than what those flood maps may tell you. It tells you that there's maybe a lot of underlying flood exposure that we have as an industry that you don't even know is sitting there and it's uninsured. And you see that when you look at that mortgage performance, that it tells you that exact same thing that we've been talking about. If you have uninsured flood risk, it increases your likelihood of default on a mortgage and your increase in overall delinquencies. And so again, there's probably a lot of underlying flood exposure inside of folks' portfolios. There was a lot more than I was expecting to see when we started looking at that data as well. So you've got climate in general, you've got property insurance costs in general, then you've got this other flood component that's another risk attribute that I think would be worth folks checking into as well. Well, especially because we see like, oh, one in a hundred year flood, and then it happens three times in like five years. And you're like, that's not a good model anymore. I don't know, that doesn't work. It's just been exposed so many times. I mean, North Carolina over the last couple of years, We've seen entire kind of towns, unfortunately, wiped away. We saw the extremely unfortunate events down there in Texas. I mean, it's time and time again. And again, it's one in 100 year for you. But when you look at a national portfolio, you've got those exposures taking place all of the time. And a lot of it's connected to those hurricanes as well, where they come up through and you get the wind damage early in Florida and Texas and Louisiana. But then you get this severe rainfall pressure that happens more inland in places that aren't used to it. And it causes massive flood risk. This is so interesting. I could talk to you guys all day about this because to me, it is fascinating. Thank you both for these insights. Clearly, climate risk is no longer just an abstract concern. Larry, as you said, it's not just something that's going to happen in the future. For listeners who want to learn more about ICE's climate data or the mortgage monitor findings, where should they go? You want to tell them, Andy? I can tell. For the mortgage monitor specifically, you can just Google ICE mortgage monitor. I could give you the URL, but it's probably easier just to Google it. So if you Google ICE mortgage motor, you can go out there and find all of our latest reports and findings. Yeah. And just ICE.com forward slash climate or pretty easy email address. First name dot last name and ICE.com. Shoot me a note. Happy to talk to you. Appreciate both of you guys. Thanks so much for the conversation. And we will talk again soon. Awesome. Thank you, Sarah. Appreciate it. Thank you. Bye bye. Thanks for listening to Housing Wire Daily. If you haven't already, we'd love for you to take a minute to rate the show and leave a comment. And make sure to tune in tomorrow for more news and insight. you
