IB Talk

Narrator 1: [00:00:05] Welcome to IB Talk, the leading podcast for the insurance industry across the United States, brought to you by Insurance Business. 

Narrator 2: [00:00:14] This episode is presented in partnership with Munich Re. For the latest episode of IBA Talk, we are joined by two experts from Munich Re to discuss inland flood risk. They delve into the recent strides being made in flood modelling and how improvements to data will ultimately allow insurers to provide more sophisticated products and solutions to consumers. 

Bethan: [00:00:45] Hi, everyone, and welcome back to IBA Talk. The Insurance Business America podcast. I'm Bethan Moorcraft, senior editor at Insurance Business, and this episode is focused on inland flood risk. Now, flooding is certainly top of mind in the wake of the absolutely devastating Hurricane Ian in late September. Shortly after that, Congress extended the National Flood Insurance Program until December 16th, 2022, providing critical financial support to flood exposed properties. And in recent months, there's been enormous amounts of innovation and product development coming out of the private flood insurance market, all with a common goal of reducing America's flood insurance gap. There is just so much going on in the flood insurance arena, and that's why today we are very lucky to be joined by two experts from Munich re America who will dive into some of these trends. It's my pleasure to welcome Inland Flood product Manager and VP Serena Garrahan and Senior Vice President Cat modeling manager James Rehbit. Serena, James, welcome to IBA. 

Serena: [00:01:46] Hi, Bethan. Thanks for having me. 

James: [00:01:47] Thank you very much. Glad to be here. 

Bethan: [00:01:50] It's great to have you both on the show. So, James, I'm going to come to you first. What flood trends are you seeing today? 

James: [00:01:58] So the peril of flood has been around for many years and we've had many significant flood events in the past. But recently what we've noticed is there's been an uptick in flood loss, both from a frequency and severity standpoint. Currently, it's very common to see flood and what I would consider non traditional low lying flood areas. This year alone, for example, we've seen major flooding in areas like Saint Louis, Kentucky and Yellowstone National Park. From a scientific standpoint, this can really be attributed to two major factors. One is rising sea level, which presents itself in significant tidal surge in the aftermath of major hurricanes. And the second factor is the preponderance of more frequent extreme rain events, which is really due to climate change. And that's what's causing a lot of these major floods and what I'll consider non traditional areas. And with the forecast for continued global warming, this trend is expected to continue into the into the future. Serena, I know you work with many flood experts in your department. Do you have anything to add in terms of that? 

Serena: [00:03:18] Thanks, James. There's definitely been an increase in both the frequency and severity of rainfall events. Increases in inland and urban flooding can be attributed to warmer atmosphere temperatures. The warmer air is able able to hold more moisture resulting in heavier rainfall events. Another one that comes to mind is Hurricane Harvey. Harvey stalled over Texas, resulting in widespread flooding, with some areas receiving over 20 inches of rain. And in that Houston area, it was estimated that up to 70% of homes damaged did not have flood insurance. Along with an increase in rainfall. We're also seeing an upward trend in billion dollar losses as a result. We've seen $18 Billion flood loss events in the last decade compared to 15 for the prior three decades combined. 

Bethan: [00:04:08] Mm hmm. Thank you both very much. It is interesting to see and we've read about a lot of these events in the news as well. Now, James, you look at this with an analytical eye as the catastrophe modeling manager. What are some of the recent strides that have been made in flood modeling? 

James: [00:04:25] Well, I would say using commercial models to quantify flood risk is relatively new in the scope of the modeling world, especially when we compare that to the more traditional perils, say, hurricane and earthquake, which have been around for largely 30 years or so. The current modeling capabilities aren't at the level of hurricane and earthquake yet, but they are improving and they're much better than they were a few years ago. And I would expect that trend to continue and flood as a peril, as a modeling peril, presents some key issues. And one of the the main ones is data collection. If you compare, say, tornado modeling to hurricane and earthquake, largely the same data elements are necessary to model tornado as it is to model hurricane or earthquake. But with flood, there are new elements that are necessary factors like first floor elevation and the presence of basement. These are data elements that aren't necessarily important for hurricane or earthquake, but are vital to the accurate modeling of flood losses. And unfortunately, those are not widely captured within the industry, or at least if they are, they don't make their way into the modeling files that we need to produce a credible estimate. One improvement that I want to talk about with regards to flood modeling really centers around the hurricane peril. And a known white spot. And Serena had mentioned Hurricane Harvey before, but a known white spot with hurricane modeling has to do with, I'll call it non tidal surge flooding, i.e. the the flooding that occurs just because of the unbelievable amounts of rain that exist in during these hurricanes. And when a slow moving event kind of parks itself over an area like Harvey did with Houston, this was a known white spot in the modeling community and it is now being addressed where hurricane models and flood models are working together so that on the hurricane side, all elements of the event from wind and tidal surge to now the sometimes significant inland flooding are being included. With regards to the future, I would expect the modeling will dictate some improvements in data capture and data collection. And then as we move forward with more events, unfortunately there will be better model calibration and I would expect this to continue on to a level where we are with hurricane and earthquake currently. 

Bethan: [00:07:08] That's great and it certainly sounds like good progress has been made, but clearly more improvement is needed, as is always the case of models, you know. Serena, what is the impact of some of these improvements in flood modeling? 

Serena: [00:07:23] So better flood modeling is really enabling us to close that protection gap by offering innovative solutions to homeowners. Just in the last couple of years, the number of private insurers that have offered flood insurance has increased from 49 carriers to 58 carriers from 2019 to 2020. And while Flood was once considered uninsurable, better flood modeling has enabled insurers to enter that market and to provide homeowners with some alternative solutions to the NFIP, which prior to recent years was really the the only way that people were purchasing flood insurance. Flood insurance itself is not a one size fits all solution. So private insurers are offering higher coverage limits as well as some additional coverages like additional living expense basement contents and replacement cost coverage. And then on the other side of things, we also see some low limit policies that might be attractive alternatives for homeowners that live in low to moderate risk areas. Some other ways that private insurers are providing flood insurance is as an endorsement as opposed to a standalone policy. This is another great solution for homeowners. Makes the purchase of flood insurance a little bit easier for them and in most cases more affordable. 

James: [00:08:51] In the beginning of this. We touched on areas like Saint Louis, Yellowstone. Receiving unprecedented amounts of rain and unprecedented flood events. And I think there's just a natural bias to think that something that has not happened in the past could never happen. And I think that's where models and their unbiased nature can provide some insight and indicate that there's a need for for protection where a consumer might not think one exists. So I think using models in that scope to kind of identify these hidden areas where the risk might be there but has not yet occurred is something that's key. And the other piece that I would talk about with regards to models still has to get around data collection and data capture and really what clients can do in order to mitigate the risk as best as they can. Again, we've seen this on on the hurricane side and on the earthquake side where models will show the credit for a a certain way to reduce risk, say, engineered shutters for hurricanes or something like that. When consumers see the financial impact of implementing these risk mitigating techniques, they're going to be motivated to make these improvements in their homes and build a more resilient structure going forward. I think flood in some ways is a lot like the parallel of California wildfire, where it's not a peril we're going to eradicate or anything like that. We have to learn to live with the peril. There will always be fires and there's always going to be flooding. But the models can show ways to better prepare ourselves for these events when they do occur. 

Bethan: [00:10:37] That's a really interesting idea there. James is kind of using models to highlight the need for protection and also kind of understanding and awareness around resiliency. And then, as Serena said, using that to kind of develop products, etc.. So it sounds like lots of exciting things are underway in the flood insurance market. We do need to close the protection gap. Serena, what do you foresee as the future of flood insurance? 

Serena: [00:11:04] Sure. So right now, the take up rates are still extremely low for flood insurance. And education is a really important part of closing that protection gap. A lot of homeowners think that just because they live outside of a special flood hazard area, that they aren't exposed to flood. But we know that that's not true. The improvements that we're seeing in flood modeling can really help us to communicate a location's flood risk to homeowners. We're already seeing this with the NFIPs implementation of risk rating 2.0. So rather than relying on flood maps, risk rating 2.0 is using location level data that reflects the property's true flood risk. In addition, we're continuing to see the impacts that climate change has on extreme weather events, causing flooding in areas that maybe have not previously flooded. So we really need to address the issue as a whole, not only through risk transfer but also in risk mitigation. Munich Re has done some work with community based flood solutions. That's another way that we can help communities to better prepare and recover from flood events. So really going forward, I think that as the data continues to improve, insurers will be able to offer homeowners more sophisticated products and additional options to protect themselves better. 

Bethan: [00:12:24] That's great. Thanks Serena. It certainly sounds like there's great need and also lots of opportunities for development in the flood insurance market. You've given us some good insights into that today. So, James, thank you very much for coming on IBA talk and sharing your knowledge with our listeners. It's been great having you on the show. 

Serena: [00:12:42] Thank you. 

James: [00:12:43] Thank you very much. 

Bethan: [00:12:45] Thanks also to our listeners for tuning in. I'm Bethan Moorcroft, senior editor at Insurance Business, and this was IBA Talk the Insurance Business America Podcast. 

Narrator 2: [00:12:55] Thank you for listening to this episode of IBA Talk. For more from the team at Munich re visit them at munichreamerica.com. Thank you for listening to IB talk for the latest episodes. Be sure to follow us on PodBeam. You can also tune in on your favorite listening channel. Find us on Spotify, Stitcher and Apple Podcasts.