Assembly Standing Committee on Emergency Management

Welcome to the Joint Informational Hearing between the Assembly Insurance Committee and the Assembly Emergency Management Committee. Today, the Committee will discuss catastrophe modeling and how this method partners with insurance. It's been said time and time again that an to the California insurance market is that insurers can only look backwards and not forwards when it comes to evaluating risk by California regulation, insurers are prohibited from incorporating catastrophe methods into their rating formula.

Based on the recent announcements by State Farm and others, it's time to come to the table to see what changes need to be made to ensure the viability of the state's insurance market. Our insurance market thrives based on competition. Availability brings affordability, which is a monumental component for our constituents.

I'm looking forward to hearing from our panelists on how catastrophe modeling works and whether these methods would help prevent insurers from pushing the pause button. Before we start with our first panel, I'd like to hand the mIc over to Assemblymember Rodriguez, chair of the Assembly Emergency Management Committee, for his opening remarks.

Good morning. I'm Assembly Member Freddie Rodriguez, Chair of the Assembly Committee on Emergency Management and Vice Chair of the Seismic Safety Commission. First, let me thank my friend and colleague, Assembly Member Calderon, chair of the Assembly Insurance Committee, for convening this joint oversight hearing and for taking a hard look at the challenges facing California's insurance market. California is a disaster prone state and climate change will only make disasters more frequent and more costly.

In addition to our traditional investments and hazard mitigation, such as seismic retrofits and wildfire mitigation, insurance and catastrophic modeling needed to be part of our overall preparedness and conversation. Mrs. Calderon and I have shared commitment to making California more resilient to all hazards and better prepared to respond to multiple disasters.

We are also committed to managing risk in the most effective way to ensure hazard mitigation projects protect lives and property. Mitigation projects, whether they reduce seismic, wildfire or flood risk, should also be designed to ensure the availability of insurance and be helpful to ratepayers. I look forward to today's discussion and hearing more from our partners in the Administration on how we are using catastrophic modeling to make smart hazard mitigation investments.

I also want to remind everybody we will be having a hard stop at 11:00 a.m. And before we go on to our first panel, I know we have two of our colleagues here if they want to say a few words before we get started.

Thank you, Mr. And Mrs.Chair.

I'm also grateful for the panel. Today, I look forward to hearing from the various stakeholders and the Insurance Department to understand how these catastrophic models can be used to better assess risk and to improve the health of the admitted insurance market, because I think that's the goal we have here, is to have a healthy market where people can have affordable rates and stay in the market. As we know, the fair plan is a temporary but not a permanent solution.

And so I look forward to a robust conversation today. Hopefully, we can outline what the path is to implement these models in a reasonable timeline. So thank you and look forward to the conversation.

You're welcome, Assemblymember Ortega.

I also would like to thank the chair for today's hearing. Looking forward to having a robust conversation and being able to ask some questions in regards to consumers and addressing our climate change, but also making sure that we're protecting our constituents and they have access to equitable coverage. So looking forward to today's discussion. Thank you.

All right. Thank you very much. So with that, we introduce our first panel, catastrophic modeling. With that, let's first call up our first person. First, we'll hear from Dr. Roger Greenier, Senior Vice President of Consulting and Client Services, Verisk. When you're ready, sir.

All right. Well, good morning. My name is Roger Grenier, and first I want to thank Chairperson Calderon, Chairperson Rodriguez, and the members of the Insurance and Emergency Management committees for inviting me to speak to you today. In my remarks today, I want to provide a brief overview of catastrophe modeling with specific application to wildfires. And I'll just note that I read the background documents circulated ahead of time ahead of this meeting, which is quite well done.

And so I hope to amplify some of those points and add some additional color to the discussion. So if we can move to the next slide. Perfect. So first, why do we need catastrophe models? Well, insurance covers many kinds of events. So for events that occur with high frequency, for example, theft, vandalism, broken pipes, insurers have abundant historical data, and they can use this data to price these types of events and also understand the range of losses or size of losses that can occur.

For infrequent events like floods and wildfires, insurers do not have sufficient data. They don't have enough data to accurately price the risk. But importantly, there's not enough data to understand how large the events can be. And so without a good understanding of the aggregation of risk across many policies, insurers are unable to manage their exposure, structure their reinsurance contracts, and really be able to ensure that they are able to pay the claims following a catastrophic event.

So for these low-frequency, high-severity events, insurers use catastrophe models to simulate the range of events and losses that can happen. So the catastrophe modeling framework is shown there in the graphic. The terminology is a bit different than what was in the background document, but basically, there are three components. There's a hazard component that describes the type of events and the intensity or impact for each event.

In the middle there is the engineering or the vulnerability component that describes the damage to different building types at different levels of event intensity. And finally, a financial component that describes the monetary loss to not only the insurer, but also the losses that are passed on to reinsurers or retained by the insured. So I'll describe the first two components in a bit more detail in the following slides. So if we go to the next slide, please. Thank you.

So, starting with the hazard component, we build what we call a model catalog, which really contains the collection of plausible, simulated, but realistic events that form the basis for the loss estimates. In the case of wildfire,in order to create the catalog, we need a good understanding of the conditions that can lead to wildfire formation in each simulated year.

We need to model not only the weather and climate conditions for that year, but also the weather and climate of preceding years that influence the vegetation and the fuels available to burn. In the background document, there was reference to recent periods of drought and high temperatures and the influences that that's had on recent wildfire risk. And so these are exactly the types of cycles that we consider when we build the catalog.

We also need to understand the variability in the vegetation, in the fuels geographically, which has a significant impact on the propensity to burn and on the fire spread. And so we use available data on fuels to capture the current state of the landscape. I'll point out here that we use historical data from roughly the last 40 years to simulate realistic weather and climate cycles, which allows us to account for the climate change that has already occurred when estimating the risk in today's climate.

So the risk today is really the default view of risk for insurance applications. But I'll point out we can also use future projections of weather and climate to simulate the impacts of climate change. And so the modeling framework is flexible enough to accommodate that approach. So if we go to the next slide, please. Thank you.

So, starting with these environmental conditions, the simulation needs to consider a number of other factors, including historical ignition locations, the networks of roads and power lines that could contribute to new ignitions. As we know, most events occur as a result of human activity, and proximity to roads is a contributing factor. We also need to understand the density of fuels, as mentioned, and the location of the wildland urban interface, where we have a mix of natural fuels and manmade structures.

We include something called urbanicity as an indicator of where people live as it relates to fire suppression. We explicitly model fire suppression activities by considering the availability of water supply and the concentrated firefighting efforts, which occur in proximity to areas with significant property development. So we consider all of these factors. Go to the next slide, please. The model also includes physically realistic fire spread.

So for each simulated event in the catalog, we consider the physical environment in and around the ignition point, and then simulate plausible weather conditions for that location and the season of the simulated fire. So the goal is to create physically realistic burn perimeter. And at this stage, the modeling software effectively intersects the property database, for example, an insurer's portfolio, and then calculates the damage to the property within and around the perimeter. So, slide six, please. Thank you.

In modeling physically realistic fires, we need to account for variations in intensity and the different mechanisms for fire spread, including surface spread, surface-to-canopy transition, canopy spread, and branding and spotting spread by embers, which is critically important in urban and other developed areas. Go to the next slide, please. So that's really an overview of the hazard component. So, moving now to the engineering or vulnerability component.

This is where we relate the fire intensity to the building damage, and we develop what we call a set of damage functions which are specific to different building, different structure types. So in addition to accounting for updated building codes, we can explicitly include structural and site-specific features that influence wildfire risk. For example, we can consider defensible space or roof type or other features that can differentiate one property from the next.

We can also consider community level efforts when this information is available, like Firewise USA, for example. So maybe it's useful at this stage to step back and recall kind of the modeling framework introduced earlier. So essentially, we've outlined the basic steps of the process for a single event. So first, we define the event in terms of its ignition location and environmental conditions. Next, we simulate the fire spread to calculate the perimeter and the intensity within the perimeter.

We then intersect the perimeter with the database or the map of insured and uninsured locations. For each location within that perimeter, we apply the appropriate damage function to calculate the damage to each property, and then we carry the damage function forward into the financial component to apply policy conditions and calculate the various loss perspectives. So that's the process for one event. And effectively, the model is just repeating this process over and over again for thousands of events on the catalog.

So we arrive at a loss reach event. And that's how you determine the probability distribution of losses, exceedance, probability curves and the other metrics that were described in the background document. So, next slide, please. So, before I summarize, I wanted to highlight this image. It might be a little bit tricky to see in the back, but it illustrates how catastrophe models can really augment historical data, particularly in areas that have experienced significant development over the years.

So these images are from San Diego, county, and we've overlaid a historical fire perimeter, which is shown in the red line, if you can make that out. So that's a historical fire perimeter, and it's shown on an image on the left around the time that the fire occurred. So this is from May of 1994. The fire was 1993. This is an area in San Diego County. So on the left is basically the level of development at the time of the event.

On the right is an image from the same area with the same fire perimeter. But the image is taken about 20 years later. So you can see the significant level of development that has occurred within the fire perimeter in that 20-year period. And obviously, at the time of the event, the loss would have been very low. If the event were to recur, it would be a very different story.

So I think this illustrates what I think we all understand intuitively, and that is that the historical loss record will have gaps, and catastrophe models can help fill those gaps to provide a more complete view of the risk. And finally, my last slide. So this slide really just summarizes what I've described today. So the purpose of catastrophe models, which is really to augment the historical record for these very low frequency, high severity events. The models provide realistic simulations of potential losses.

By considering climate, weather fuels, historical sources of ignition, we consider physically based fire spread and account for the influence of the built environment in addition to accounting for wildfire mitigation, fire suppression, and importantly, as we've just shown, the models capture new development and increased exposure over time. And finally, I'll just say that catastrophe models help advance a goal that I think we all share, which is to improve our understanding of an ability to manage wildfire risk.

And as we work towards this goal, we contribute to helping individuals, families and businesses recover from these devastating events. So with that, thank you. I'll take any questions.

Thank you very much. We'll head on to our next panelist and then our next person here next, we'll hear from Ahmad Wani, Chief Executive Officer and co-founder of One Concern. When you're ready, sir.

Thank you for having me. So, One Concern is a seven-year-old technology startup based in Menlo Park. We are not deployed in the United States. Our first market was Japan, and so most of our work has been with Japan. So I'm going to be building on sort of what Roger has already given you is like a detailed understanding of how catastrophe modeling works and how it functions. Right. So, as a quick overview, go to the next slide, please.

So, One Concern is trying to build what we call as the digital infrastructure for physical risk. We do three things. Mapping, analyzing, and monitoring every piece of the world's built environment and understanding its connection to the global economy. So we don't necessarily build new catastrophe models or opine on climate change or those kind of things, but essentially we work with like 10 or 15 sort of these climate or catastrophe modeling companies to ultimately layer them on what we call as the graph of the built environment.

I'll try to illustrate the problem we are trying to solve through an example. So, 2017, when we were still kind of getting started, we got a phone call from an insurance company. They were trying to win the account of one of the largest logistics companies, which you all know. And they were like, can you guys just, you do something around AI? And we saw your video on the Stanford engineering website. So can you guys come and impress them?

They said it in a nicer fashion, but we went there and we ran 10,000 scenarios on this one facility in Seattle, this largest logistics facility on the West Coast. What we found was that at intensities equivalent to almost a six magnitude, 6.6, 6.3, 6 .4 magnitude, they lost power at 7.1, 7.2, they lost water at 7.8 the nearest bridge was shut down for inspection, cutting off access to the port. And it was only at a nine magnitude that the building itself started to get started to shake.

What we started to realize that most of these buildings, which are Fortune 500, Fortune 1000 facilities, they're not in flood zones, they're not vulnerable to wind, earthquake, wildfire, nothing, right? But the surrounding infrastructure is the vulnerability. In fact, in 98% of the cases, the surrounding infrastructure is much more vulnerable and gives up first. Right. So next slide, please. We call this risk, the dependency risk, right.

It's the risk not just necessarily to the facilities, but physical risk, which is going to affect the bottom line of these companies or the operability of homes, also of commercial facilities. So it could be massive power outages, supply chain disruptions, et cetera. Next slide. So essentially, we started to wonder, what are people doing about this? And what is the size and scope of this risk? 20 years back, dependency risk was negligible. In the US 40 years back, it was zero.

We do not really have any massive power outages due to weather or supply chain outages due to weather. In the last just 20 years, this dependency risk has gone up 10 times. In the last almost 37 to 40 years, almost two orders of magnitude, 90 times. Flood risk has only gone up double-digit percent over that period. So what's the glitch here? The glitch is the infrastructure budget deficit.

We continue to spend a lot of money, especially the commercial sector, to upgrade their own facilities, the industrial facilities, the large residential towers, but not enough in upgrading our infrastructure, the surrounding infrastructure and there's a huge infrastructure budget deficit, whether it's the railing power grid, the road systems, the highways, et cetera, the water systems, the sewer systems.

So, essentially, the existing way of understanding these catastrophe models is GIS systems, which is essentially layering them on a two-dimensional map, like a flood map, and seeing who's in the flood zone today, who's going to be in the flood zone 10 years from now, 20 years from now. We take a slightly different approach. We were like, well, the world is not just two-dimensional, it's multidimensional. So we should model the world as a graph. Think of it like a spider's web of nodes and edges.

Every facility is a node, every other facility is another node, and they're connected through these edges, and those are those dependencies. The road systems, the highway systems, the internet cables, the telephone cables, the water pipes, power lines. And we set out to build this graph by collecting all of this data, one by one, for each city. That's why it took us seven years to even launch our products. So we collected hundreds of cities worth of data.

And then we start to use machine learning to be able to sort of understand, AI systems to generate the data for those cities which did not have this data, because now we knew how cities are built and how they will respond to events. Now, we got lucky, because in 2019, we had an opportunity to go to Japan. I had an opportunity to go and present there, fast forward from there, four months from there, I was presenting on the floor of the Japanese Senate.

And why it resonated with them was because Fukushima had the exact same characteristics, like that logistics facility. It was not inundated by the tsunami, it was actually standing tall, but it had a power outage, and it started to get overheated. All they had to do is to get in, hook it up to a generator on the bottom floor so that it cools down. But the road access was blocked. So the two dependencies there were power and transportation.

So essentially in collaboration with the Ministry of Economy, the Ministry of Land and Infrastructure, the Prime Minister's Office in Japan, along with the insurance company Sompo, we launched kind of like a mega project over many years to be able to build this system, to be able to build this digital twin for Japan, and then essentially work with our cat modeling partners to be able to bring in those cat models and break the digital twin and see those ripple effects.

Because now that we actually disturb the digital twin, whether it's shake a node or break an edge, we now see ripple effects, 2nd, 3rd, 4th order effects. We measure everything in time, so we can measure sort of power downtime, which this facility is going to have today, or 10 years from now or 20 years from now due to climate change, or supply chain downtime, or highway downtime. Let's say the digital twin in the US has 8 trillion data points.

It maps out every electric pole in the US, every crane in the port of Oakland, every substation detail. Now I want to quickly talk about a few applications of this. Obviously we work a lot. Our main market is sort of insurance and financial services. So we work with companies to be able to help them price this into securities, whether it's bonds, derivatives, asset management, et cetera.

We also help insurance companies reduce their business interruption claims, which they are not able to otherwise have a good grasp on because they're understanding the facility directly. But a major application has been mitigation, climate adaptation and mitigation. And I wanted to just focus on that specifically by pointing out two case studies. One of them is actually from Japan.

So we have this concept of a resilience score, which we have built from the Ministry of Economy of Japan and the Sompo Insurance Company, which is now being rolled out to some of those folks. Now, obviously every facility is affected by downtime differently. It's important to understand that, let's say a hotel, even half a day of people not being able to check into the hotel, they will lose cash flow, right?

But a data center, nobody is visiting the data center every day, so they are more bothered by the water to cool the data center, like the nuclear power plant. All of us like technology companies, like financial services, we can work from home for a year without any cash flow impairment. So everybody has a different sensitivity to downtime or physical downtime before they start to impair cash flow. And we use those to generate exceedance probabilities and ultimately come up with a resilience score.

I want to point out that we've built a resilience score for a business, and that's what we really do. We build it for FedEx, a Walmart, a Dow chemical, and those kinds of people. But at the same time, we started to build this for the government. Now, the government resilience score is very different, right. And it's important to sort of understand that the government resilience score has three components. Life safety, which is essentially, people shouldn't lose their lives, and we don't want any casualties.

Second is the whole livelihood angle. The time it takes for the city to recover quickly, whether it's power, transportation, et cetera. And then the third is, the city shouldn't default on its bonds, et cetera. And when we are trying to mitigate, or when those folks in Japan are trying to mitigate, they are basically figuring out what those KPIs are today for both life safety, you know time it takes to recover from the major lifelines, as well as the economic standing of the city.

Then once they decide where they are today and where they want to be, let's say, for instance, one of the cities we launched with was Kumamoto in western Japan. There, the mayor ran on flood agenda. He ran on flood and he was like, I want to. And he got elected, right? And so his thing was, nobody is going to die because of floods in Kumamoto going forward. He was like, zero casualties in Kumamoto going forward.

So his entire mitigation plan or budget is being spent not in terms of insurability, for instance, but primarily in terms of providing recommendation incentives for homeowners to upgrade, as well as evacuation routes, as well as early warning systems. And then we are able to sort of monitor that and see, okay, these are the places that you can actually get your KPI back up in three years. You will have that zero fatality type situation, obviously, for a certain type event.

Now, if you're dealing with, obviously, insurability, if you want to focus on that, then you have to do a different type of recommendation. So the recommendations would be different based on what those KPIs are today and where you want to be. Right. So that's how the resilience modeling framework works. Where we basically want to understand from our in-quotes client or whoever we're working with, sort of what those KPIs should be. And then we can make the relevant recommendations.

A bridge may seem very weak, but if it leads to nowhere, then there's no point upgrading the bridge. If places like paradise or other places are already now greatly affected does it make sense to spend a lot there? Because it's kind of like already really decimated. Wildfire risk is pretty low. It's like Silicon Valley banks, like the safest bank. Right. So it's already affected. So we need to be very thoughtful and data-driven in terms of how we make recommendations as to who we mitigate.

Like, just because some city does not have grant writers shouldn't be not receiving grants. So we need to have a more data-driven approach at the state level so that we can understand our KPIs in terms of life safety, in terms of insurability. And then we can sort of be like, okay, these are the recommendations. And it could be a different water pipe, a different power line, a different wild and urban interface mitigation, et cetera.

I just wanted to end-

Uh-oh

Emergency. Emergency.

That's kind of power downtime.

Okay.

Yes, we're good. It was only for a second. All right. I guess our backup systems do work here in the cabinet.

Yes.

So with that, sir, if you could wrap it up. Got one more minute so we can get on to some questions.

Yeah, so I just wanted to end with a quick video, basically. What this. You can go to the middle of the video. Yeah. Great. So essentially this is kind of like how our system works. This is just like a screen grab of the system. And the way it works is this facility, for instance, can you just pause? Perhaps you can just pause at any point.

So essentially what I'm trying to demonstrate here is, and there are a few more slides which we don't really have time to get into, but I wanted to showcase, like, for instance, as an example, a fire station in Marin County, which is completely new, but it's going to have seven days of power outage and it's going to be having no access from one and 101 in an earthquake. Right.

So essentially we need to be able to actually mitigate in a more system wide, holistic fashion so that these systems, these fire stations can actually perform their function during an emergency. Same thing with like a Stanford hospital, for instance. It's the world's safest building. Right. It can survive a 2500 year hit earthquake, but if it doesn't have water after two days, its operating room is going to be dysfunctional. Right. And we have to think about this at a system wide level, at a holistic level.

So thank you so much for having me.

All right. Thank you very much for your presentation. I'll open up some questions. Rebecca, you had a question.

Thank you so much. I want to thank the chairs for welcoming us to your Committee hearing and for having this. The question of how my constituents get covered for fire insurance is something they think about on a daily basis. My entire district is Wildland Urban interface in the East Bay of the Bay Area and I really appreciated the presentation that you had on catastrophe modeling. My neighborhood is a firewise neighborhood. We are taking all of those steps.

My fire chief has done incredible work to put shaded fuel brakes in, to put sprinkler systems that are automatic out into the Wildland space so that if a fire were to start, there would be mitigation immediately. And yet we can't get insured. And the answer today is the state plan. Well, we talk about not allowing for catastrophe remodeling because it would raise rates.

But if right now my entire community is having to go on the fair plan, which is exorbitantly expensive, I guess my query is, if we allow for modeling that acknowledges the risk that my community faces, also acknowledges those mitigation factors that they're taking seriously, and that will help them. I mean, my gut tells me it would be cheaper than them all going on the fair plan.

I also don't think the fair plan can stand up if we put every single Californian who's at risk on the fair plan. So that's a separate issue. But I guess that's the question I'm trying to ask, is I know that most states allow for catastrophe modeling, and where does the impact on rates relate to the current cost of fair plan? I assume it's somewhere in between current rates and the fair plan, but I wanted to ask that question. Do you know?

I'm sorry, the question about

How this affects rates, the catastrophe modeling, do you know, or should we save that for the second?

Well, I think it will vary on a case by case basis. I think your point is well taken. So there's two sides of the coin. You need to understand sort of the underlying risk, and then you also need to understand what steps have been taken to reduce that risk, whether it's at a homeowner's site or at a community level.

And that's where the catastrophe models, because they're data-driven, because they can take advantage of new data and iterate over time, gradually improve the view of risk that comes out. That's where the models can be very helpful. The steps that you described in terms of the mitigation features definitely is not included in historical data. So really, the model helps kind of augment the historical data because it's a data-driven approach.

And interestingly, when you showed the historical data on that San Diego neighborhood. One of the things I've learned from my fire chief who tells me that I don't live in the right place is you know things like roads also change the fire. Right. So there, where you had no roads because there was no community, there's more houses. But your risk also changes for how far that will go with natural fuel breaks that are built into communities.

So I don't think historical data is really problematic because there's so many factors that change as we build communities. And so I just wanted to be here to really thank the chairs for holding this hearing, for having this conversation. I know my community is eager for us to have a marketplace that does acknowledge the risk that they face, but also allows for them to live in the community that they love so much. So thank you.

Sorry. The one thing to sort of understand is there is very limited reinsurance available for wildfire, if any, for California. Right. Most reinsurers have pulled out of wildfire just nationally. I mean, obviously these insurers are providing insurance in other states, wildfire insurance in probably 16 states, but they are holding it on their own books. So it's sort of important to understand that the other thing is wildfire risk is not just the same as flood risk or earthquake risk.

There's no actual wildfire risk map right now in the world. Right. Because it's a very dynamic map. You know, it needs to be seasonal. It needs to be changing continuously with mitigations. Like if now we suddenly have more shrubs growing somewhere else, it needs to be changing. Flood risk and earthquake risk are not that dynamic. Right. Wildfire mitigations, similarly, like $1 gives you back $4 or $10. We can debate on the earthquake side, but that's when the actual earthquake happens.

That could be within one year for a location. It could be 500 years in a different location for wildfire. You need to spend that $1 probably every other year. Right. So my point in saying this is if you do provide, I'm all for providing catastrophe modeling capabilities and allowing for them and for rate increases, but it's not necessarily because we've done a fair bit of mitigation. It's not going to be like an asset-level pricing anyways, because the models are not that granular.

You're going to see very blurry models for wildfire risk, and so the mitigations will not be reflective of like, okay, you did this, so you're going to suddenly have a reduced rate. That's not going to be necessarily like they're going to change the rates for the entire state. Now, obviously it might reflect for certain communities, but the science is not where it is for understanding earthquakes or floods. I can certainly go into more detail.

The presentation you had reflected that mitigation. So I assume it does take into account your catastrophe modeling mitigation. Is that not true?

Is that a question for me?

Yeah, sorry. Because he's saying that you wouldn't be able to take those mitigations into account. Well, because to your point, a shady fuel break needs to be maintained. Right? I think that's your point, right? I mean, the perimeters that we now have ordinances in my cities that you have to have those perimeters from home. So it's no longer optional. It's now the law because of our risk.

And so I guess where we have those laws on the books, I would assume you'd be able to take that into account because now, you know, our fire chief, he was out driving my neighborhood the other day. So you know that we're maintaining those mitigation standards and hopefully that would be reflected in rates.

Yeah, I would say two things. Yes, the short answer is yes, the mitigation would be reflected. I'd make the point that there's a level of mitigation that can occur at the property level, like, for example, defensible space that can have an impact. But there's also in some cases maybe a more important level of mitigation that happens at the community level that prevents the fires from getting into the communities in the first place. So I'd say both go hand in hand.

And again, the framework we described is a data-driven framework. So as this information becomes available, sort of a key process here is let's capture that data, let's make it available so that it can make its way into the models. And gradually over time, you improve the models and that leads to just better outcomes, I think, in the long run.

Pilar, did you have a question?

I did. Thank you so much. Really appreciate this hearing that the chairs brought together. And this is a huge issue in my community as well, I represent San Fernando Valley and Santa Clarita Valley, both surrounded by mountains and hills and lots of brush. And in my community specifically, we've been very heavy hit by this issue. I live in a condo community of over 700 units.

We got dropped from our fire insurance a few years ago, couldn't get insurance, then got catastrophic insurance, which is wildly expensive, have had multiple $1000 assessments on our units and they raised our HOAs $200 a month on top of that. So we have seniors in our community with fixed incomes. I don't know how people are affording it, and it's a huge, huge issue. So I'm trying to understand practically how catastrophe modeling works.

Like, in our community, it is guaranteed that there will be fires in the hills around my neighborhood every single year, 100%. However, in the 60 years that our community has been there, none of the buildings have ever burned from a fire. Right. They've always been able to get it out. The fire department is close enough, et cetera. I know you can't do on-the-spot modeling for my neighborhood, but just kind of trying. Can you walk me through how catastrophe modeling would work in this instance?

Because I am not sure if it would be better for our community or worse for our community and how this kind of plays out?

Yeah, I think that the short answer would be, so I mentioned in one of my slides, I use this term urbanicity, which I'm not sure is a word, but it basically means trying to understand what's sort of the land use in and around the perimeter of the fire. And one of the things that you see when you look at historical burn scars is that you do see these pockets where you have a fire that advances to a point and then it stops.

And generally it's because of fire suppression efforts. So that's one of the things that we try and capture in our models as we're propagating the fire, recognizing in that sort of simulation process, where do you get into a different land use, where you've got more urban population, so that you then presume some level of fire suppression to protect those areas. So that's generally how we do it. I'll say that it's a tricky process because it's not strictly driven by just the physics of the wildfire.

Right now, we're also trying to model sort of the human behavior. So that, again, goes back to the data and just having better understanding of the firefighting capabilities in communities and then building that into the model. So that's generally the approach, is that we do recognize that fire suppression happens and is part of the outcome of these simulated events.

And so we really are looking for ways or have incorporated ways, but always looking to improve ways to make that more robust as we develop these views of risk.

Assembly Member Jim. You had a question?

Yeah. First of all, thank you. I appreciate this, and I very much appreciate data-driven models as we try to assess risks, and I'm disappointed that these catastrophe models are not allowed by law here in California because I do think over time, with the proper data, we might be able to get to a point where insurers, instead of using a chainsaw, can use a scalpel in how they carve up areas based on risk because right now, for example, I have a man who owns an inn.

He's right on the coast. He can see the ocean from his inn, yet he's considered by insurance company a super high risk, even though oftentimes when in the morning he's got rain, but he has forest around him, and so if there was catastrophe model that showed, 'hey, this isn't really a high risk area,' he might actually be able to afford insurance for his inn.

And I also liken it to--in 2014 in my district--then Secretary of Natural Resources John Laird was at the Coyote Dam in Mendocino County in February and the Army Corps of Engineers was releasing water and they were releasing water because they had a book that said on this date, 'if there's this much water, you should be releasing water.' The problem is there was no more rain that year, and so we released a bunch of water that didn't need to be released.

Fast forward, we're now using atmospheric river research data to be able to predict when we should be releasing and when we should not be releasing water, otherwise, we'd still be back releasing water at the wrong time of the year. So the idea that we as a state cannot get our arms around using catastrophe modeling is ridiculous from my perspective. So thank you for being here, and I hope that at some point we can change the law.

Thank you. With that, Assembly Member Gipson, you had a question?

Thank you very much. I want to say thank you to Chairwoman Calderon and also yourself, Chairman Rodriguez, for bringing us all together to address this very important topic that's before us. I don't represent a wildfire area. I represent Watts, Compton, Carson, Wilmington, North Long Beach, Harbor Gateway, Harbor City, Wilmington, and San Pedro.

We don't have this, but it's all of our problem in the State of California, and so with that being said, the catastrophic modeling is critical to reflect the real risk insurers are facing in this State of California, but from what I've heard, it is one piece of a larger fix in this huge dynamic. The inability to include reinsurers in the cost and the delay in the processing time also needs to be addressed. It needs to be addressed. Did I say it needs to be addressed? It needs to be addressed.

And are the Department of Insurance and insurers committed to working to remedy this situation that we are faced with because it is doomed to get worse? And so that's my question because, again, all of California, whether you live in a wildfire area or not, all of us are in this thing together, and we're seeing the ripple effects by insurance companies not writing insurance moving forward. So if you can address that, then I may have a follow up question. Depends on your response to that question.

So I guess the question is--

Processing time?

Yeah, processing time for reinsurance? Well, so, I guess as you said there--

The judge wants to say something real quick. Maybe that would be a good panel, a good question for the next panel regarding insurance, maybe?

You think so?

I think so, maybe, but if you want to take a stab at it real quick, I was just thinking right now. You get the question but--

I guess as you said, it's--yeah, go ahead.

Okay. Let me ask this question real quick. What have we not asked you that you want to leave with us before you leave?

No, that's a good question. I think you've got, as I said, the background document--I think was very well done. I think it was a pretty good explanation so I feel like you've got a good understanding. I think the challenge is how do you balance and sort of introduce this change, which is always hard to do over time and I think that's probably the fundamental issue. Maybe it's not so much about the science. Oftentimes these problems aren't about the science of technology. It's about people. It's not a math problem. This is a people problem.

I think I agree 100 percent. I don't think this is 95 percent a catastrophe modeling problem. Obviously, that science has existed for some time, right? So I think that can be incorporated. It can be a more dynamic process, especially--I think there needs to be a single source of truth for what mitigations are being done, and it needs to be a clear, auditable database because insurance companies don't have the means to be auditing that. They make three, four percent a year.

So there needs to be a mechanism for people to actually sort of say, 'this is what I have done in terms of mitigation, and I continue to do for a community, for these firewise communities,' for instance, right. So all of that, but I just want to leave you guys with one thought. It may not necessarily be popular in this Committee, but obviously introducing catastrophe modeling is urgent, right, for this, but when we talk about mitigation, we shouldn't just be looking at the current top of mind issue or just wildfires or just earthquake.

We have to be looking at it from a very holistic fashion because the next event or the next challenge is not going to be State Farm threatening us that they're not going to provide insurance, it's going to be completely something different, and so we just want to have a very holistic, data-driven approach across all hazards, across all these dependencies, across all our KPIs of balancing out life safety, insurability, economic impact, and all of that.

Okay. Thank you. One next--question for the next panel. Go ahead.

Thank you. Question for the witnesses. I know we're going to be hearing from the Department of Insurance next. One of their concerns is that they believe that the modeling that would be used, the software, should be made accessible and should be transparent to the public.

I understand that raises obvious proprietary trade secret issues on your end. My question is, how has this been handled in other states, the issue of being able to examine what goes into these modelings to make sure it is fair and scientifically sound without giving away your trade secrets?

Yeah, good question. So a couple of things that we do as a company, one is that we make our documentation or sort of model documentation available to the departments of insurance and basically available publicly. So if someone wants to look at, for example, the wildfire documentation, it's a 100-page document that describes the tools and data that we're using. In terms of how other states address the issue, so there's a couple of different approaches.

So one is sort of the Department will handle a set of interrogatories and just sort of interpreting or sort of questioning aspects of the model and some back and forth where we will explain to the Department about how the model works. In other cases, there's a third party that would evaluate the models and provide--evaluate the model against a set of standards and determine whether or not the model meets those standards and make the models available in that way. So it varies depending on--

So you would be willing or you guys do make it available--the modeling--to third parties or even the Department of Insurance to examine the software?

We'd be open to lots of different ways that we could use to move this forward. We, as you say, it is important that we protect our IP, but there's lots of ways to get around that.

There is a specific way you validate a hazard or a catastrophe model. This is a very solved problem, believe me. There is a bottoms up way where you walk an expert or a set of experts through it. We have a technical working group composed of professors from Stanford, Berkeley, all those people, and also in Japan, Japanese professors. So they approve our models.

So that's a bottoms up, and then top down, you basically have to be scrutinized by, let's say, the Department of Insurance, where they'll be like, how did your model perform back tested? How did your model perform in paradise or in Napa in a certain wildfire or a certain earthquake? So, that's how--and then you have to publish all of these. These are not perfect models, right?

Have you started these conversations with the Department of Insurance? Have they started looking at your models or not yet?

I think in our case, I have been in the past five years on a few calls with the Department, so I think they have a reasonable understanding of how our models work, but we haven't got into the specifics of a review process or anything like that.

All right. Thank you. Thank you.

Okay. Thank you very much for the first panel. Now we'll move on to our second panel. Second panel is Utilization of Catastrophic Modeling. First, we'll hear from Ryan Buras, Deputy Director of Response and Recovery, California Office of Emergency Services.

Yes, and also let's bring everybody else up. Also, we'll have Glenn Pomeroy, Chief Executive Officer from California Earthquake Authority, and we'll have Lynn von Koch-Liebert--hopefully I got that right--Executive Director, Strategic Growth Council, and we'll have Mike Peterson, Deputy Commissioner for the Climate and Sustainability from California Department of Insurance, and we'll have Lucy Wang, Deputy Commissioner from the Special Counsel, California Department of Insurance. So with that, we'll go on with Ryan when you're ready.

First off, my testimony says, good afternoon. It's good morning. Chair Rodriguez and Chair Calderon and Members of the Committees, my name is Ryan Buras, and I'm the Deputy Director for Recovery Operations at the Governor's Office of Emergency Services.

Our mission, as you know, at Cal OES, is to protect lives and property, build capabilities, and support our communities for more resilient California. Cal OES: principal responsibilities include the coordination of all the state's disaster response, emergency planning and preparedness, and disaster recovery mitigation and homeland security activities. For state level operations, modeling is generally used to characterize threats and risk.

Cal OES utilized modeling efforts by partner agencies to support catastrophic planning for hazards that have the potential to impact large areas of California and overwhelm local and state capabilities.

For example, the scenario and impacts used in the most recent planning effort, the 2022 Southern California Catastrophic Earthquake Plan, reflect in-depth analysis of models and tools such as the Uniform California Rupture Forecast and FEMA's standardized system for estimating risks from disasters in the Hazards Program.

FEMA's Hazards Program provides standardized tools and data for estimating risk from earthquakes, floods, tsunamis, and hurricanes. Risk assessment resources from this program are always freely available and transparently developed.

FEMA, Cal OES, and local partners utilize data generated through hazardous modeling to estimate the impacts of disaster during the planning process so that resource requirements, gaps, solutions can be utilized. Additionally, the location and flood hazards identified in the Northern California Catastrophic Flood Response Plan or the Flood Insurance Rate Map floodplains, part of a national insurance system maintained under the National Flood Insurance Program, NFIP.

During the transition from response to recovery, we found during recent flood events that some of the hardest hit areas weren't asking for assistance quickly enough because they were so heavily impacted by the event. Cal OES in coordination with FEMA, use synthetic aperture radar, SAR, data from the ICEYE Flood Insight product to identify these locations and to ensure that all areas where significant flooding was projected using this model were inspected for potential impacts.

This data allowed us to identify areas where the need was great, but this ability for communities to reach out to us for help was limited and we were able to ensure preliminary damage assessments were completed as quickly as possible. SAR is a remote sensing technology that utilizes radar signals to gather information about the Earth's surface and has recently proven to be a valuable tool for emergency managers to understand flood depth in the extent during disaster response.

By acquiring SAR data at regular intervals during the recent flooding event, FEMA tracked changes in flood extents. This actually included the depth of the water at each parcel. Not only could FEMA determine what structures were in or near the zone of threat, they were able to detect the depth of flood structure was likely exposed too.

This data aided in understanding the progression of flood, identifying areas of risk at further inundation, and it also helped us with the advanced planning so we can move resources to those in most need. Cal OES also used data from the 2022 Esri Market Potential data set to support our request for the presidential declared disaster. The insurance data plays a critical role in evaluating a county's recovery potential from a disaster operating at both the county and our ZIP code level.

It provides valuable insights into the level of coverage within the county, helping us understand the extents of how prepared that community is. For the Ferndale earthquake and subsequent events, we have utilized the Esri Market Potential data set, which includes the variables: quote, 'have any homeowners or personal property insurance' and quote unquote, 'have home insurance coverage for earthquake or flood,' unquote. Several local, state, and federal hazard mitigation system programs use Model N as a factor that determines whether a grant project is eligible or not.

It is not used to rank or score projects. In addition to the Hazardous Program, the main modeling tools used for hazard mitigation efforts include the FEMA Benefit Cost Analysis, BCA Toolkit. FEMA requires that all hazard mitigation projects must demonstrate that benefits will outweigh cost in the FEMA-approved methodology using the FEMA Benefit Cost Analysis Toolkit. As I stated earlier, Model N in this process is only used to verify eligibility, not score applications or make decisions or project approvals.

The hazardous program provides default data sets for inputs that any community can quantify risk regardless of data availability. The State Hazard Mitigation Plan is a document that considers the intersection of expected losses from a natural hazard impacts to equity priority communities and climate change factors that affect the hazard impacts. Cal OES also implies vulnerability criteria to determine eligibility for a Prepare California Match program, which can cover non-federal cost share associated with the FEMA Hazard Mitigation Program.

And finally, Cal OES utilizes fire hazard severity mapping performed by our partners at CAL FIRE when we determine eligibility and prior visitation for the California Wildfire Mitigation Program that we have a joint powers authority with our partners at CAL FIRE. Counties supported by this program were selected and ranked by identifying the overlap of high wildfire hazards and also their high social vulnerability risk at that parcel and census tract level. Thank you for your time, and I'm available for any questions after the contestants.

Thank you. Next we'll hear from Glenn Pomeroy, Chief Executive Officer, California Earthquake Authority. When you're ready.

Good morning, Assembly Members. Thank you for inviting me to be here today. The CEA is a state's not for profit earthquake insurance enterprise established by the Legislature in the mid 90s in the wake of the devastating Northridge earthquake. We're publicly managed but entirely privately funded, writing only residential earthquake insurance and only in California. Now, with over one million policyholders, CEA is the largest earthquake insurer in the United States and one of the largest in the world.

Fortunately, large, damaging earthquakes are infrequent, but unfortunately, when they occur, they can be catastrophic. Because earthquake insurers lack access to historical data upon which to project future loss estimates and establish rates, the use of models for this catastrophic risk is both necessary and authorized. CEA is required by California law to set actuarily sound rates, that is, rates that are not inadequate, not excessive, and not unfairly discriminatory, and an additional requirement is that CEA rates be based on best available science.

To meet these twin legal requirements, CEA does use catastrophic models to number one: develop rates and mitigation discounts to determine the amount of claim-paying capacity that we need for our policyholders to establish CEA losses or estimate CEA losses after an event and for ongoing strategic planning purposes.

Our enabling statutes contemplated model usage for establishing rates and laid out the requirement that rate differentials may be established only if the information and methods are consistent with the state of the art of seismic knowledge and within the scientific community. Following inception in 1996 and once the portfolio of insured homes had fully transitioned to the CEA, the CEA Governing Board began examining the appropriate amount of claims-paying capacity for the portfolio.

Catastrophe models were utilized for key metrics such as the capacity needed to provide coverage for target levels as well as the capacity for repeats of historic earthquakes. To this day, and with the governing board support and approval, CEA contracts with three widely commercially recognized cat loss modelers: Verisk, CoreLogic, and Risk Management Solutions, RMS.

We use this to view a broad range of modeling results for planning purposes, including evaluating claim-paying capacity targets, as well as to determine the expected reinsurance recoverable for the many risk transfer contracts that we purchase. A unique feature of the CEA statute is that it exempts CEA from the California Insurance Guarantee Fund, CIGA, and instead contemplates that CEA would pay claims on a pro rata or installment basis if an event exceeds our claim-paying capacity.

CDI regulations require that upon the occurrence of an earthquake event in the state that is likely to give rise to significant losses, CEA staff shall create a scientific model of anticipated losses within seven days. Here again, CEA will use cat models to satisfy that requirement. Another feature of the CEA enabling statute is the required hazard mitigation discount. CEA is required to provide a discount of at least five percent if the insured mitigates their home, but the maximum discount we now currently make available is up to 25 percent.

The schedule of discounts was developed from catastrophe model output. We also use cat model output for strategic planning and evaluating a variety of 'what if' scenarios ranging from catastrophe model response drills to a coverage design, and finally, all other California earthquake insurance writers can and do use catastrophe loss models for setting rates as allowed by CDI regulations.

The same section also allows private insurers to use catastrophe loss models to develop the estimate for the Fire Following Earthquake Risk, which is part of the homeowner insurance rate development process. So thank you, Mr. Chair. Happy to answer any questions, if there are any.

Thank you. Now our next person be Lynn from the Executive Director, Strategic Growth Council. Lynn, when you're ready.

Good morning and thank you, Chair Rodriguez and Chair Calderon, Assembly Members; appreciate you inviting me to join you today. My name is Lynn von Koch-Liebert. I'm the Executive Director of the California Strategic Growth Council. Now, Strategic Growth Council, or SGC, is a ten-member council that is comprised of seven state cabinet agencies and three public members.

Our grant and loan program seek to make an impact in a variety of issue areas, from agricultural land to affordable housing to racial equity, but all prioritize technical assistance, capacity building, community engagement, and multi-benefit projects to advance healthy, thriving, and resilient communities.

We administer a variety of social and physical infrastructure programs that are relevant to today's discussion, including the Affordable Housing and Sustainable Communities, which invests in multi-unit affordable housing and sustainable transportation infrastructure, the Transformative Climate Communities, which invests in place-based multi-benefit projects to achieve environmental, health, and economic benefits in disadvantaged communities, and the Regional Climate Collaboratives, which funds local cross-sectoral collaboratives to build capacity to access funding for climate resiliency projects.

Now these programs prioritize climate and equity, which requires a forward looking and accessible approach to investments. SGC's programs do not currently use catastrophic modeling in their funding decisions. Instead, we ask applicants to use Cal-Adapt and other tools provided by the state, along with community engagement practices, to identify risks of a changing climate in their community and to share how their project builds resilience.

Cal-Adapt is a free tool developed by the State of California's climate change assessments that makes climate data much more accessible for the state, community members, and local governments to utilize in climate adaptation, resiliency, and planning. For example, one program SGC administers that utilizes Cal-Adapt is the Community Resiliency Center or the CRC program.

CRC is a new program that invests in the planning and building of a neighborhood level resiliency center to provide shelter and resources during climate and other emergencies, as well as services and programs that build community resilience year round. Last month, CRC released their first round of funding in which 98.6 million dollars became available to communities across the state. CRC requires applicants to use Cal-Adapt's Local Climate Snapshot Tool, which provides climate projections for temperature, rain, and wildfire to determine risk exposure at the project location.

We also ask applicants to build on Cal-Adapt's data, with best available qualitative and quantitative data sources, inclusive of regional and community data, to discuss gaps and needs for local priority populations. Applicants should also use community engagement methods to ensure they are providing direct, meaningful, and assured benefits that address potential climate hazards.

Applicants are then scored according to how they leverage their findings to mitigate environmental, social, and economic risks. They must explain adaptation and resiliency measures taken to reduce these risks through their project design.

This includes program services, emergency activation processes for resiliency centers to respond to climate emergencies, as well as specific design features for the facility and surrounding CRC campus that can appropriately and effectively respond to identified climate hazards. Overall, SGC aims to address climate-related risks through its programs, especially for disadvantaged communities and other priority populations facing the brunt of the impacts of climate change.

We use tools like Cal-Adapt to support communities that are most vulnerable to the impacts of climate disaster, build social and fiscal climate resiliency because they are impacted by economic, health, and other intersecting issues, and as we consider the use of existing and new data-driven resources, we should be intentional and careful to avoid exasperating environmental justice issues by further disinvesting in these areas.

Instead, the tool should inform how we can invest and to support communities at highest risk through physical and social infrastructure. SGC takes this equity-focused lens to invest in the social and physical infrastructure to make climate hazards less damaging in the future and easier to recover from in the present. Thank you and I'm available for questions.

Thank you very much. Next, we'll move on to Mike Peterson, Deputy Commissioner for the Climate and Sustainability from the California Department of Insurance. When you're ready, Mike.

Thank you very much. Good morning, Chair Calderon, Vice Chair Essayli, Chair Rodriguez, and Vice Chair Waldron, and Members of the Committees. Thank you for the invitation to testify here today. I am Mike Peterson and I serve as Deputy Commissioner for Climate and Sustainability at the California Department of Insurance under the leadership of Insurance Commissioner Ricardo Lara. The overarching approach for Commissioner Lara on addressing the challenges posed by wildfires for insurance has been to continue to prioritize reducing risks to California communities.

The Department of Insurance has focused on a multi-year effort on engaging with consumers and stakeholders as it assesses how new tools can improve risk management, make residential and commercial insurance more accessible and reliable for Californians, and maintain competition and ensure stability in the state's insurance marketplace.

The Department of Insurance has been very clear that benefits to consumers are of utmost importance as we strive to increase the availability of reliable insurance from the admitted market, ensure the long-term stability of rates, and incentivize the accurate recognition of wildfire mitigation efforts. The following actions provide further examples of our multi-year effort to improve risk management.

In 2019, Commissioner Lara convened his Climate Insurance Working Group with environmental advocates, researchers, wildfire officials, and insurance experts, which focused on producing recommendations to reduce the increasing impacts from climate-intensified wildfires, flooding, and extreme heat, and to close insurance protection gaps.

Furthermore, in December 2020, Commissioner Lara held a virtual public meeting focused on fire science to examine how wildfire mitigation could reduce the spread and risk of future wildfires and the extent to which wildfire catastrophe modeling, aligned with wildfire mitigation as a rate making tool can impact market competition and insurance availability.

The Commissioner invited testimony from 14 individuals with wildfire expertise and experience, including the Institute for Building for Business and Home Safety, three researchers from the University of California, representatives from four catastrophe modeling firms who have developed models for insurance, two actuaries with California specific experience, a member of the Cultural Fire Council and Yurok Tribe, a community wildfire mitigation specialists, and representatives from the Center for Insurance Policy Research at the National Association of Insurance Commissioners and CAL FIRE.

Then, in 2022, Commissioner Lara promulgated his groundbreaking regulation that recognizes and rewards wildfire safety and mitigation efforts made by homeowners and businesses, a regulation that became operative in October 2022. The fourth California Climate Assessment was published in 2018 and found that if greenhouse gas emissions continue to rise, the frequency of extreme wildfires will increase and the average area burned statewide would grow by 77 percent by 2100. This finding underscores the importance of risk mitigation actions and informed the work of the Climate Insurance Working Group.

In 2021, the Working Group released the first ever Climate Insurance Report. Among the recommendations, the Working Group stressed the importance of risk assessment to insurance availability and reliability. One recommendation advised the insurance commissioner to hold public meetings specifically on the use of catastrophe models as a tool for estimating catastrophe losses, noting that a public meeting would provide the public an opportunity to discuss the use of such policy tools.

The use of actual historic catastrophe losses has been the long standing method used over several decades for estimating catastrophe adjustments in the California rate approval process. However, historical losses do not fully account for growing wildfire risks or risk mitigation measures taken by communities or regionally as a result of local, state, or federal investment. Expanding insurance options for Californians and promoting community wildfire risk mitigation were foundational pieces of this report's recommendation.

Investment in natural infrastructure that mitigates risk at the community level, such as forest management, coastal dune and marsh restoration, as well as prescribed burn strategies to reduce future wildfire severity, are essential to community risk reduction and could be reflected in modeling tools.

I will conclude by emphasizing that the discussion on catastrophe models as a tool is part of a progression of public meetings and engagement with the risk assessment expertise of universities and stakeholders, as we have evaluated the benefits to consumers. Thank you for the opportunity to be here, and I would be pleased to take your questions.

Okay, thank you. And last, we'll hear from Lucy Wang, Deputy Commissioner and Special Counsel from the California Department of Insurance. When you're ready, Lucy.

Hi. Good morning. Thank you. I'm Lucy Wang, and I serve as Deputy Commissioner and Special Counsel to Commissioner Lara. Thank you for the opportunity to speak with you this morning. As you just heard from my colleague, with the continued threat of climate change and its associated effect on our global and state economy, insurance companies are continually trying to manage their risk, and the California Department of Insurance is also continually trying to explore additional tools that could benefit our consumers through increased insurance coverage options.

To that end, the Department sent a notice last week regarding its workshop scheduled for July 13th regarding catastrophe modeling and insurance. As you heard from my colleague, this July workshop is a continuation of an effort that Commissioner Lara started after taking office. He continues to be on the forefront of addressing the risks associated with climate-intensified wildfires, and this workshop supports his long-term strategy of protecting consumers while maintaining competition and ensuring stability in our state's insurance marketplace.

This is a priority not just for the commissioner, but for the entire State of California, and this July workshop is going to focus on exploring the legal questions presented by the use of catastrophe modeling in insurance rate approval. As Mr. Grenier had set forth earlier in his testimony, there are several potential benefits of wildfire modeling for consumers.

Wildfire catastrophe modeling has the potential to increase accuracy in the rate making process by using real-world, on the ground conditions to simulate wildfire behavior and to estimate wildfire risks.

Catastrophe modeling could also recognize the positive impact of home-hardening investments, community level mitigation strategies, and it fuels reduction projects at the regional and local levels. Coupled with a clear commitment by insurers to write in more high wildfire risk areas of our state, especially for those who are vulnerable and living in the WUI, catastrophe modeling could create greater rate stability and sustainability. However, these models are subject to important transparency requirements and other consumer protections under current California law.

The Insurance Code requires that the Department be allowed to evaluate any such models to assure that rates generated from such models are fair, justified, and accurate, and the goal of our July workshop is to discuss how we can incorporate catastrophe modeling into California's insurance rate making process within our regulatory framework. The specific questions that the Department seeks to ask and discuss are one: the significance and impact of Insurance Code Section 1861.07, which requires catastrophe models to be publicly disclosed.

Two: what potential methods can be employed to ensure that the Department can review catastrophe models in accordance with Section 1861.07 while preserving the proprietary information of a particular catastrophe model, and lastly, other states have incorporated alternative methods of structuring model review, such as the use of an independent third party or third party panel, and are there methods used by other states that would be a viable option here in California?

Meanwhile, the Department is also working to evaluate the potential risks and necessary safeguards that should be put in place for catastrophe modeling to be used in our rate making process. The Department intends to host a second workshop later this summer which would address the practical considerations associated with catastrophe modeling, such as one: model stability, should the Department expect property insurance rates to be more stable if insurers were permitted to model wildfire peril. What is the impact of catastrophe modeling on insurance rates for our California consumers?

Two: model accuracy. Over the past few years, catastrophic wildfire events have occurred that have allowed insurers to support significant rate increases based on adjustments in the Department's rate making formula. Will permitting the use of catastrophe modeling more accurately support rate requests by insurers in the future? And most importantly, how would allowing catastrophe modeling benefit California homeowners and businesses?

Would the use of catastrophe modeling accomplish something that couldn't be done through some other means, such as a regulatory adjustment to how we currently use historical catastrophe losses? The goal of these workshops is to continue the commissioner's commitment to exploring how new tools such as catastrophe modeling can make residential and commercial insurance more accessible and reliable for Californians as we protect consumers and create safer communities. Thank you for the opportunity to address this important topic here today, and I'm available to answer any questions.

Thank you very much. With that, we'll bring it up for Members for questions. Any questions from anybody? Assembly Member Irwin, you have a question?

A couple of comments too. I really appreciate you putting this hearing together and sort of new in the insurance area. I don't sit on either one of these committees, but with the new district that I have, we have a lot of issues with fire insurance. The biggest, I know that the Insurance Commissioner is very focused on reducing risk and rate stability. Our problem is that people are being dropped and they are unable to be reinsured in our area.

And I appreciate we've done a number of workshops with you, but we have homeowners associations that cannot get insurance. Luckily, the FAIR Plan has increased its coverage to 20 million, which is helpful for some of these townhouse associations, but this is a huge issue.

I have a staff member who last week was--a couple of weeks ago--was told by farmers that she is not insured anymore and is going to have to figure out some other insurance plan, even though our entire area is considered high risk, even the homes that are not right near open space.

So you have your work cut out for you, and hopefully the Legislature can help along, but I don't know how much conversation there has been--but--about the earthquake authority and does it make sense to have a fire authority? So maybe that's something that's been discussed in the Insurance Committee. What does that look like, and does it make sense for the state to start looking at that as a possible solution? Partial solution.

Question for me?

Yes.

Well, thank you. The CEA is a creature of statute, assigned the responsibility of ensuring for earthquake in 96, doing so, also establishing a mitigation program. We've now retrofitted 20,000 homes and hope to double that in 2019. We were given the responsibility to oversee the Catastrophe Wildfire Fund so we take these responsibilities seriously and we discharge them.

This question has not been brought to us, and frankly, we have our hands full dealing with the same market challenges that private insurers do in terms of reinsurance capacity shrinking and getting more costly, so we're focused on that and are discharging our mission, which is hopefully a polite way of saying we have not looked at that, and don't feel it's within our statutory assignment to do so.

Right. I'm just talking about a separate California fire authority because obviously the situation that we're in right now is untenable so I guess maybe it's something to be explored.

Definitely something we can look at then. Okay. Any other questions? Ortega. Assembly Member Ortega, a question?

I'm also new to the Insurance Commission, but I've been interested in kind of the announcement that was made by the insurance companies. We've been getting a lot of questions from some of my constituents. I don't have a big fire area, but I do have some, and some have actually called and said they're not being renewed.

So just, I heard the Insurance Commissioner staff talk about balancing the competition within the insurance market with protecting consumers and making sure that we look at what's happening. And just looking at one of the spreadsheets here, I'm just looking at the losses, right, and the increase going up from 1997 to 2018 at 614 percent is what I'm looking at.

So how is it you talked about the four kind of layouts that you're looking at modeling and how that's kind of your framework? So can you talk a little bit more about what that means in terms of what we're seeing and moving forward, and how can we be supportive?

Well, holding meetings such as today are very helpful in terms of the opportunity to discuss this important topic. I think the question that you are looking at in terms of what your constituents are facing with non-renewals is a little bit separate from this workshop today, and the Department's workshop is specifically focused on incorporating catastrophe modeling into our rate making process.

I think the issue, which is also a difficult one regarding your constituents and their ability to obtain insurance and non-renewals is kind of separate, and I'm happy to discuss that further at a different opportunity, but specific to the points that I discussed in my testimony, we really are looking at how we can incorporate catastrophe modeling into our rate making process.

One: given the fact that we have insurance laws in place that require a balance between transparency and we're balancing that with these modeling companies proprietary information, and then two: what are the practical realities incorporating rate making?

Number one, for a Commissioner, is always how we protect our consumers both in terms of, if we allow such a tool, is it going to increase availability of policies, what is the cost of that, and those are the things that the Department seeks to look into with these workshops.

And the second point I would just make is that insurance regulators across the country are dealing with the issues of climate intensified events, whether they're fires or floods or storms. And so when it comes to non renewals and issues, we do monitor those. We do data collections. And our role has really been to try to help people navigate what those options are and how to encourage insurance companies to expand how they're writing in our state. Thank you.

Thank you. I appreciate you breaking it down into two parts. Thank you very much.

Thank you. With that, Assemblymember Shiavo, do you have a question?

Sorry, I'm back here. So what I understand you saying is that it sounds like the Department of Insurance is moving towards catastrophe modeling and figuring out how to incorporate that in California. Is that right?

That is correct.

Okay, excellent. So my question is, catastrophe modeling, I feel like now, is being put out here as the holy grail of where we need to get to in California. It's going to change everything. Do we have any assurances, commitments? We have insurers who are pulling out. We have reinsurance that's disappearing. How confident do we feel that this is going to make the change that we need to see happen in California, that folks will come on back, that folks will not leave, that people will reinvest?

Do we think that this is going to be this kind of silver bullet to solve the problems? I mean, it's still California. We have wildfires. It's a costly state to be in. There's real challenges around it.

Sorry. I'm going to talk to the mic.

Its okay. I won't be offended.

Thank you for your question. I think that to take a step back, catastrophe modeling is a risk assessment tool. That's one piece of a bigger puzzle, right? There's the risk assessment, there's the risk reduction. And then what Commissioner Lara is very interested in is a firm commitment from insurance companies to expand writings, especially in areas where they have stopped writing in the wildfire risk areas. And so to your question, is this, to what degree is there a silver bullet? As I mentioned earlier, insurance regulators across the country are dealing with the rising costs from a number of global issues.

And so I think that what's important is that we really get started on the risk reduction and at least looking at the risk assessment tools like we are doing with this upcoming workshop so that we have all those pieces in place as we look at the overall situation. But just to reiterate, we've seen severe wildfires increasingly in California. Our own climate assessments are looking at the future, and that is the challenges that we face. And so there are those different parts.

And I think that the previous regulation on safer from wildfires really tried to focus on what can consumers do to reduce their risk at their home. And then this upcoming workshop is looking at what are the risk assessment tools, and will insurance companies commit to writing more and expanding their writings with kind of different risk management alternatives? So thank you.

So, for the risk reduction. I mean, we've heard Assemblymember Bauer-Khan was here earlier. I don't know if you heard her comments. They've literally put sprinklers in the forests around housing and they're still not getting coverage, right? And so they're going to pretty, what I would feel like are pretty extreme steps to make sure that they're reducing risk and still not getting covered. I've heard just anecdotally from other folks that they've done everything that they were supposed to do to be able to bring risk down on their property.

And still insurance companies have not lowered their rates. So that's kind of one concern I have around how far that will go. And then I just wanted to put a plug in here because I think you mentioned some kind of review process when catastrophic modeling is used, because I'm not sure if you heard earlier, but the example that I used of my community, which butts up against the Santa Susanna Mountains, which literally have a fire at the end of my block every year, maybe multiple times a year, but it's very contained. It's out in a half hour.

And in the 60 years in our community, none of the houses have ever burned from it. Right. And so I heard from the panelist earlier that they try to incorporate the suppression tactics as part of the modeling, which try doesn't really make me super confident about how that works out. And so I think having people be able to review or take input or kind of have a broader process than putting something into a computer and spitting it out and assuming that that's the most accurate kind of model that you could come up with.

So I just want to flag that as a concern of mine. But how do we address, I feel like it's such a hard balance that we have right now, right? Because insurance companies are taking on huge risk and expenses and it has to pencil out for them, right? To be able to be in this business here in California. At the same time, our constituents either can't get covered, can't get their rates down, or just are paying so much for it. It's unaffordable. So how do we address if people are doing all of this risk reduction and mitigation and then they're not seeing a result from their insurance company on that?

And is there a plan B? Because we don't know that all of this is going to work out. I don't know. I don't know. Maybe you know, but I don't know. And I don't know Jackie's idea. Are there ideas that we are coming up with to think about a plan B? I'm really concerned about this issue. It literally comes up every time I'm in the district. It's a huge, huge issue in our district and throughout California.

I feel like we need to be moving to action that really makes an impact in people's lives and understand we have to proceed cautiously and get a lot of information through that process. But there's a lot of concerns. Thank you.

Thank you for your questions and concerns. And to respond to that, first off, this is a really important issue. I wanted to kind of just note two things. And one is that when it comes to risk mitigation, we as a state have accelerated our approach to that over the last couple of years, but many of those efforts haven't yet come to fruition. So I think we are in sort of a moving situation, and I'll just flag two key points on that. One is that Commissioner Lara's safer from wildfires regulations just became operative in October 2022. We received rate filings in April.

And so, consumers moving forward will be required to receive incentives based on those home hardening factors, but also whether you're in a fire-wise community and whether you're in a fire risk reduction community. And so that is still yet to come to full fruition. And then secondly, as a state, we've appropriated significant new amounts of money towards wildfire risk reduction. But that does take time to work itself into projects that reduce risk and can be reflected in a model or in an insurance view of the situation.

And then we've also seen a dramatic increase in the number of fire-wise communities over the last five years. I think it's gone from somewhere in the 200s to nearly 500 very quickly. And so as the example that you noted, those fire-wise communities are getting set up, they're doing more things, and they're going to be doing more things that I think are best practices moving forward that will have more and more impact. And so I think you've got kind of the implementation of regulations, the appropriations going into projects, and both of those are going to have an impact on risk. Will it be the ultimate silver bullet? We'll keep learning and keep working from that.

And then to your point, I'm sorry, this is weird because I have to speak in the mic. You're sitting behind me. But then to your point, I mean, what you're raising are the practical and difficult questions that we seek to address in these workshops, which is catastrophe modeling is perhaps not a silver bullet, but is billed as having the ability to provide greater accuracy when it comes to providing fair and accurate relates.

And so what the Department really wants to explore is what does that mean for our California consumers? What is our ability to assess how those models work? And number one, if we allow that as a risk assessment tool, are we going to get a commitment from insurers to write in our WUI and our high risk areas? So those in the upcoming months, those are very much the issues that the Commissioner seeks to explore in these workshops.

Okay.

Thanks. And with that, we have a question by the Chair, Assemblymember Lisa.

Yes. Thank you. So just a quick question. Following your workshops, how quickly do you think it would take for the Department to allow for cat modeling? I mean.

That's a tough question.

I know it's not a gotcha.

No, I know it's not a gotcha. And that's essentially what we're exploring. I mean, what's different about California than the other states is we have a regulatory framework in place that requires the balance between using a tool such as catastrophe modeling with our right to transparency. And I think that was touched upon.

So I'm encouraged by the presentation made by the first panel in terms of their openness to making that information available to us, because that's what we're going to discuss in the first workshop, is, unlike other states, our statutes require a level of transparency that other states don't require. And how do we thread that needle? And so I think in terms of how quickly we can make it happen really is going to be dependent upon the information we learn and our ability to balance our rate making process with our regulatory framework.

Thank you.

Okay with that. Assembly Member Essayli, you had a question.

Thank you, Mr. Chair. And thank you for being here. It's a really important conversation. I just want to better understand the hurdles that we have to overcome to implement this. Let me ask, Mr., is it Pomeroy?

Do you guys currently use this catastrophic modeling to write earthquake insurance policies?

Yes, sir.

Yes, we do.

Okay. Are you guys subject to the same insurance code that was referenced 1861.7?

No. Mr. Chair and Assembly Members, there's a specific body of law that created the California Earthquake Authority that contemplated the use of models. Given the fact that earthquakes are so rare, there isn't historical data to rely on. So earthquake riders must use catastrophic models, and we have since inception.

Okay. And then, Ms. Wang.

Let me ask you. So this statute that requires all the information to be made available to the public, is this something that the Legislature enacted, or is this required by the Proposition that created you all?

Yes.

I'd be happy to have that legal discourse with you, but no. Yes. The Legislature created the statute when it enacted Proposition 103 35 years ago. So it's part of that framework. And so, I think the issue is just addressing how we can allow catastrophe fee modeling as a rate making tool in light of the fact that we require transparency.

Do you think it would be helpful if we adjusted the statute to kind of mirror what the Earthquake Authority has for purpose of a catastrophe modeling? This allows the Department... I do think it's important that you have the ability to review the catastrophic modeling with your experts and make that determination, but also balancing the intellectual property that goes into making these models. And it's a business that's how they make money. So it is a balance. You have an oversight duty, which I know you guys take very seriously.

So, I think you'd have a very receptive group of legislators if you guys were to say, this is something we want to do, this legislation or some change in the law would be helpful. Is that something that you guys could consider?

No, absolutely. I think you make a great point, and I think that is the type of enthusiasm and encouragement we need as we enter these workshops and we look more closely at exactly how much information modelers are going to provide us so that we have assurance that the use of these models is going to be done in a way that will benefit consumers first and take into the practical considerations of how we assess that. So if we go through this first workshop and it arises that perhaps there needs to be more of a legislative fix, then I'm so glad to know that we will definitely work together with our legislatures to make that happen.

Well, I won't speak for my colleagues, but I know I would certainly welcome any legislative proposals you guys have that would streamline your ability to evaluate and approve these, because I think there's hard questions and then there's easy ones. And to me, this is an easy one because every other state, to my knowledge, uses these and the Earthquake Commission uses them. So I think there's a lot of data out there on how they work and how they're helpful in the insurance market to better assess risk.

And I think to kind of respond to some of my colleagues questions there. I mean, if we had better risk models, we know which properties are actually at the high fire risk, as opposed to these kind of generic and wide maps that are being used now, which aren't very accurate, then we would maybe have a smaller group of people in the FAIR plan, for example, rather than large groups of homeowners.

And then maybe HOAs who aren't really in a fire risk can get insurance instead of right now being knocked out. So I do see a huge benefit to using these models and having better, accurate risk assessments. And I would just sort of echo the Madam Chair's sort of questioning there on a timeline. I know you guys are working diligently, but the sooner we can do things, the better because we don't want to lose more insurers. I think we have to maintain a healthy market. So with that, I will yield. Thank you.

Thank you.

Okay, we have another question by Assembly Member Ortega.

Thank you. So going back to the catastrophe modeling and that being kind of the direction we're headed, while I appreciate that other states are already using this, I do feel like California is the fourth largest economy in the world, and we're a leader in many spaces. And so I wouldn't be hesitant to rush into something we haven't done because we haven't done it for a reason. So I guess one question that's still kind of the elephant in the room is would using catastrophe modeling prevent insurers from pressing pause or issuing new policies?

Thanks for your question. It's, again, it's a risk assessment tool, so it's just one piece of what goes into it. Ultimately, insurers decide where they're going to write under what risks that they will. They do currently use risk scores, wildfire risk scores, when it comes to charging certain policyholders more in high risk areas versus low risk areas. But this is a risk assessment tool. So as we explore options through this workshop, we are looking for a firm commitment from insurers that they will expand their writings. But ultimately it comes down to that risk assessment ability to reduce the risk and then attracting insurers to have good competition that will make insurance more affordable and reliable.

Okay, thank you. With that, Assembly Member Irwin, your question.

Yes, thank you. Just to follow up on Assembly Member Schiavo's comments and the response, the home hardening and having these fire-safe councils is great, but what we have found is that insurers, when you ask for the discount, have said, well, we're not going to be insuring in your area anymore. So it is critically important that we maintain that healthy insurance market. But, following up with Assembly Member Ortega's comments, I think more data is better and knowing where the real risks are.

But you are also taking some of the less risky properties out of the equation. And now you're going to have many homeowners in extremely high fire zones that I would imagine when it all comes out, won't be able to get insurance at all. So how would we, I assume that that's been part of the discussion. Right. Once you do your modeling that you find some homes are in extremely precarious situations.

I think that when it comes to catastrophe modeling, insurers are already using some of those models for their underwriting. And so I think that some of that is already being used when it comes to the rate approval, rate making process, I think that you've got, again, these two things going on. As we get more hardened homes in the state, that does reduce risk, and as we do more community mitigation, that does reduce risk.

So that's on the risk reduction side. On the risk assessment, the catastrophe models may more accurately determine what risks are there today compared to historical data. And so that can affect the decisions on where an insurance company writes or where they don't write. But really, I'd say that that risk mitigation, that home hardening, it has a stronger and stronger impact the more homes in a community that are hardened.

And so as the wildfire mitigation grant program ramps up, and as these regulations require insurers to provide incentives for that, we do expect to see more hardened homes. And for that to have an impact, regardless if you're in the highest risk zone or the more moderate, it's still going to make those communities more insurable. And so I think just really the straightest path towards insurability runs through that risk reduction. And so we're really trying to put as much focus on that as possible.

Just a follow up, just on the California Wildfire Mitigation program, we have the Brace and Bolt program. Many homes have been mitigated, and you can do that in a way that is not as condensed as what you just mentioned. And we are looking at that. That is the reason we're mitigating whole communities, hopefully to buy down the risk of that whole community so that community can get insurance.

Okay with that, I just have two quick questions, as a lot of this has kind of affected a small community in one of the areas I represent. Currently, right now, is there anybody responsible for regulating rates based on fire mapping?

I'm not sure. Could you clarify what type of fire mapping?

Regarding fire mapping risks, is there anybody that regulates or oversees that on how they increase the rates based on the map?

So we review and scrutinize all the rate filings that come in. Insurers typically have been using what they call wildfire risk scores, and our recent regulations do make those scores more transparent to the public. The companies that create the scores, they're not insurance companies, so they're not regulated by the Department. But on the case by case of each rate filing, those scores are coming in with the rate filings, and the Department does scrutinize and review those prior to approval.

Okay, and then just one last question once again to the community that's been affected in my district. Is there a way we can protect homeowners? Because I know there's good actors and bad actors right in the industry, and some of these bad actors, I think, take in their own hands fire mapping and regulates how they increase the rates. Is there a way we can kind of. What protection is there?

Maybe, I know you're having some workshops coming up that protect that homeowner because they're getting these bills just overnight that. Okay. As effective today, your rates are quadrupled. Right. They can barely afford what they have now. This happens. And in some cases, these folks may be out of a home because they can't have any way to pay for that increased cost of the risk. Right. So I don't know. Maybe as you can look through these workshops, you're having the protection for the homeowners that get these things unexpectedly. Then worst case, they're out of a home. Right?

Yeah. No, I really appreciate your question. Just to further clarify that when rate filings come in, they are scrutinized and insurance companies have to justify why they're charging what premium to which series of customers. This is something that we continue to look at. Given the last five years of severe fires, there's lots of losses in many of the insurance companies experience. And so that does have an impact on what rates they can justify in their rate filings.

But I appreciate your question about this community, and I think that it's something that we'll continue to really look at. And we do hope that by rewarding the homes that are able to do defensible space and harden their homes, that they will help to some extent. Thank you.

Okay. Thank you very much. So with that said, we'll move it on to our last panel, and I'll turn it back over to our Chair, Calderon, for our last panel. Thank you. Our last two panels.

Thank you, Mr. Chair. And I'd like to thank the panel. Good discussion. So the next panel will discuss wildfires and catastrophe modeling. And on this panel we have Daniel Berlant, acting State Fire Marshal of CAL FIRE, Frank Frievalt, Director of the Wildland-Urban Interface Fire Institute at Cal Poly San Luis Obispo, and Nancy Watkins, Principal and Consulting Actuary at Milliman. Whenever you're ready, sir.

Alright. Well, good morning, Chairs Calderon, Rodriguez, Members of the Committee. Daniel Berlant, acting State Fire Marshal, as well as CAL FIRE's Deputy Director of Community Wildfire Preparedness and Mitigation. Today I'm going to talk to you a little bit about Calfire's wildfire mitigation efforts, a little bit about our risk modeling and hazard modeling and how it affects what we're doing, and then a little bit of what we've done to work with the Department of Insurance and other stakeholders to really advance these mitigation efforts in California and make sure that they are meaningful. In the past couple of years, California has made major investments in wildfire mitigation efforts that have been a crucial part of our strategy in combating wildfires across the state.

And really thanks to multi-year funding by you, by the Administration, we have been able to really significantly increase assisting communities in the mitigation efforts that have been talked about today. Defensible space, home hardening, community level mitigation, as well as forest management. In fact, you have invested over $2.7 billion in just the last couple of years in this effort.

And during that time, we've taken a significant amount of that money, in fact over $500 million, and awarded that to local fire safe councils, local communities, to augment the state's efforts to really empower and enhance the fire prevention and forest health projects that we believe are critical in reducing our fire risk, as well as restoring our forest back to their natural state.

Now, California homeowners play a critical role in fire prevention through the maintenance of defensible space through home hardening, which I'll talk about in a moment, as well as kind of community-based mitigation efforts. And back to the funding, thank you to increased investments in staffing, we have been able to increase the number of defensible space inspections, the number of educational visits that we can make to homeowners living in wildland areas to help them create that work around their home.

In fact, we've increased by 67% in the last, compared to a couple of fiscal years ago and been able to increase to performing over 300,000 defensible space inspections. And this work is critical because this is the clearing of dead or dying vegetation that is flammable around homes, increasing their chance of surviving.

But in addition to that important work, we have really focused in on how we build homes and where we build homes, ensuring that homes are hardened, meaning that the construction materials that they are built out of can resist the embers from wildfires. And so that combination of good defensible space and hardened of those homes really does increase that home's chance of surviving.

But while in the last 15 years or so, we've had some of the most stringent building codes in the world, here in California, in wildland areas, many homes were built before that code. And so that is where retrofitting and again, this term of hardening, those homes becomes so important. But let me just tell you, our wildfire risk reduction efforts that we are doing are working.

Our post-wildfire research and data has showed us that homes that had defensible space or that had hardened their homes or built out of the current building code materials had a significant increase in chance of surviving a wildfire. So really, to summarize, our wildfire resilience strategy has really been a nested approach. We are focused on the home and the community. We look at the wildland urban interface around it as well as the forest, and making sure, again, that the ecosystem is healthy.

But in order for us here in California to be successful in these mitigation efforts, alignment is critical. We have to make sure that these core elements are the same, that other industries, such as insurance industries, local ordinances and fire departments, that we are aligned in what standards we are asking for. Because if we are all telling or requiring something different, it makes it very difficult for the public to know which of those matter, which of those are important.

And so alignment on these efforts has been a huge focus of ours over the last couple of years. My colleague Mike Peterson from the Department of Insurance talked about our effort together, the Administration, several departments, including CAL FIRE, working with the Department of Insurance to lend our expertise in wildfire mitigation to make sure that statewide standards for home hardening and community hardening are taken into account when it comes to insurance availability and insurance incentives. And out of this partnership came the safer for wildfires framework.

And it really is a document for insurance companies that really focus on both parcel level and community level measures that can be used to create a consistent approach for the insurance sector and empower those communities to reduce their wildfire risk before a disaster strikes.

And what is so important, though, about this framework and why I really want to acknowledge how important this framework has been is that it aligns the best practices that we require at the state, the minimum requirements that the state sets in wildfire mitigation. But it takes in the latest best practices, the latest science, the latest research, and ensures that those insurance companies are using that same framework. And so, again, the key to that framework is alignment.

We all have to be doing and saying the same thing. And I know my colleagues here will probably hit on it as well. Many of the people that have testified in front of you today participated in that effort. And that really has been, I think, an important part of this strategy is all stakeholders coming together, agreeing upon that science, agreeing upon those factors.

So I'll take a step back and talk a little bit about our efforts in both hazard modeling and I'll focus on hazard, but then talk a little bit about risk. And there is a difference between those two. CAL FIRE's involvement in hazard modeling most recently has been the updates to the fire hazard severity zones. This is a map of state responsibility area as well as in local responsibility areas that take in hazard conditions. And again, hazard and risk sometimes get used interchangeably.

And risk models that insurance companies use are different than hazard modeling that we use when it comes to the mitigations that we require. Hazard really is based on the physical conditions that create the likelihood and the expected fire behavior over really a 30 to 50 year period without considering the mitigation measures such as home hardening or even a recent wildfire. So again, these are long term impacts.

Risk, on the other hand, is really the potential for damage that a fire can do in an area using existing conditions, but also accounting for those modifications, such as fuel reduction, defensible space, ignition resistant building material. So the mapping that maybe, Chair, you referenced early on, I really want to be clear that that is hazard mapping. We are making long term decisions in how we build homes when we map these areas. We are just simply using hazard factors.

These maps are developed using a science based and field tested model that assigns a hazard score based on factors that include the influence of the likelihood of fire and the fire behavior that we likely will see in that area. Now, often I get asked, but I live in a subdivision, why am I in a fire hazard area?

And the model that we're currently using, based on better science, really is able to take into account extreme weather conditions, specifically wind, that can pick up embers that are produced up to a mile away and carry those embers into more urbanized subdivision areas. And we have seen fire after fire. The North Bay obviously hit some of the hardest in the last five years, where they are urban communities that are still impacted.

And so those areas still are going to receive a level of hazard even if they don't have Wildland right next door to them. Some of the other factors, though, in hazard that we take into account include the vegetation type. Obviously, the thicker the vegetation, the taller the trees, the more embers they're going to produce, the more significant fire behavior you're going to see. And so flame lengths, embers.

As I mentioned, the terrain, typical fire weather in that area are all critical components of our overall hazard mapping, which again, is hazard modeling. But switching gears a little bit into wildfire risk modeling. Over the last couple of years, our office has been engaged with a number of stakeholders, including fire chiefs, insurance experts, and many others, to develop a list of recommendations for you on how to understand wildfire risk mapping and wildfire risk when it comes to communities and individual homes.

And I'll just point back to Assembly Bill 642 by your colleague, Assembly Member Friedman in 2001, directed us to identify the wildfire mitigation factors that would need to be included in a risk model, as well as how to determine how do we get that data and to identify the barriers that it takes to actually truly create a wildfire risk modeling.

Now, a risk modeling workgroup that is made up of a number of stakeholders, again, fire chiefs, insurance industry representatives, Nancy Watkins, who you will hear from in a moment, has also been just invaluable in helping make sure that, again, we have alignment in what those risk factors look like. And so by really bringing in our subject matter experts from various fields, we've been able to work to address a statewide concern about wildfire risk modeling.

Now that workgroup is just finalizing the document, and in the coming months, we look forward to bringing something back to you that lays out the complexities that really need to be taken into account when we talk about wildfire risk. But the bottom line is that during our effort, it's been made very clear to us that it is a complex issue. We have to have alignment, as I talked about earlier, on the core elements of the factors we are including. Access to that data is critical.

And then making that data accessible to the fire service, but also to the general public is going to be an important part of the work. And so we have to make sure this information is transparent. So I will end today with, obviously the peak of our fire year is just ahead of us. We've already seen, even with all of the rain, we've already seen over 1600 wildfires across the state. And so Californians need to take steps now to prepare their homes.

I'll tell you, I've been out on the weekends doing my defensible space, making sure my home is hardened, making sure my community, I too live in a firewise community. All Californians in wildfire areas need to take the steps now to ensure that when we get to the peak of fire year that everybody is ready. So I really do thank you for the opportunity to talk about our mitigation efforts. And after my colleagues are done, happy to answer any questions you might have.

Thank you.

Alright, thank you. My name is Frank Frievalt. I'm the Director of the Wildland Urban Interface Institute with Cal Poly. Thank you, Chair Calderon and Chair Rodriguez, for hosting us. And I want to be sure, make sure I understand our time frame here. I heard we had a hard stop at 11, is that correct?

We did.

It's been extended a bit.

Yeah. We're going to try and go to...

Alright, I'm actually going to skip the slides and just get to a couple of points directly to some questions I heard. I would encourage you to review the written comments that are submitted. They're detailed and they answer many of the questions you ask today. So I'm going to drill down on a couple. In general, my position, your question is about cat models.

I'm saying "cat models and..." What that means is, you've asked some important questions about surety or accuracy or how do we do those things? From our collaborative work with cat modelers, we understand that there are several things they don't have in there that they would greatly benefit from. It would provide more of the granularity that we need. At the end of the day, whatever direction you go, follow the science. What aligns all of us is the physics of heat transfer.

If we all agree that we don't want the buildings to burn down, then it's simply a function of understanding how does that happen. And there's really three ways. And all of our work here is a response to catastrophic conflagration levels of loss of property and life. And I really commend you for doing a joint Emergency Management and Insurance Committee meeting, because these two are inseparably connected.

So the disruption of those fire pathways are three. Vegetation to vegetation within one half to one quarter mile of concentrations of 100 structures or more, with separation distances averaging less than about 100ft under the right wind conditions. That's when you have conflagrations. The second part that we need that's not available to cat modelers is an actual measurement of fire suppression capability to disrupt transition from vegetation into structure and structure to structure.

We're working on a basic set version of that now, but it's not a tool they currently have that they need, because that's a significant mitigation. If you can get the right resources on time and at the right place, you can stop that transmission. The third thing that they don't have that they need is a true structure to structure fire spread model, where structures are the primary fuel type. That doesn't exist.

Most of the models that do that substitute a known fuel model for vegetation, for a structure, and you can't count for mitigations in that. The key to this, as Chief Berlant said, most of this work has to be done by the property owners. We have situations where they can't due to means or complexity, and that's where we have grant funding, and we have other ways to do that, but to provide homeowners property owners clear guidance, to Chief Berlant's point.

We have to have alignment, and that alignment has to be around the mitigations that are evidence based to help disrupt fire pathways. And the last piece that we need is data. And here's a specific reason we don't have what we need in that space. I'm not trying to be an insurance person here, but I spent a lot of time, and especially with Nancy the last few years.

The way Chief Berlant and I have looked at this, I've been in the fire service since 79, and all of it's been in the Sierra Nevadas. We look at structure to structure spread when we get in these conflagrations. Insurance industries, the insurance industry has two pressures on it that don't really foster data sharing. One is we're concerned about some antitrust issues for good reason, and the other is market issues. Right.

You don't want a big saturation of risk in one area, so you tend to disperse what you underwrite. I believe there's been maybe a disconnect between aggregate risk of physically separated properties. If they're all mitigated properly, let's say to 100%, that's not the same as what your adjacent risk is and they can't know that right now. So it's structure to structure spread is what we're responding to. And we currently don't have that level of data spread.

So when I say the discussion sounds to me like it's going in the right direction, I believe cat models have a very important place in what we're doing. I think we also need these other elements I've discussed, and we need that part to be transparent because your local government and your state government need to know what does fire spread look like in all three of those areas. How do we disrupt it? And that tells you what mitigations you need to do.

And by the way, you don't need to do them everywhere. There are optimal places to put that. When you do flood protection, you don't put a six foot moat around a community. You mitigate in certain areas, fire pathways, they respond to things. We can model those well, so that part needs to be transparent so that we can properly manage that risk from the government side. And it also needs to be extracted and used for proprietary purposes for all the market reasons you understand.

We don't have that yet either. And the next speaker is an expert in that and will cover some of that. So to recap "cat models and..." The "and" piece has to be structured structure spread modeling. We're not quite there. Be able to measure the ability of your local fire services and state fire services to disrupt fire pathways from vegetation to structure. And we have to be able to model most probable pathways last half to quarter mile. You can't mitigate the whole forest. Right.

We need to know how to intelligently optimize disruption in that last space. Those need to be transparent so that we can use them in an open government setting, and the public can see, they can understand what does our community look like now with these? What would it look like without them? And we have to have that turned over annually. That's the way that we're going to get there. So thank you for your time. Appreciate that.

Thank you.

Thanks, Frank. I'm Nancy Watkins. I'm very grateful for you all having me here today. I'm a Principal with Milliman in San Francisco and an Actuary. As an actuary, my profession is focused on accurately measuring and communicating risk. My practice, we consult for insurance companies, including most of the major ones in California. We also work with the National Association of Realtors.

We review cat models and help state insurance regulators come up with mitigation schemes that are fair. We've reviewed models for FEMA, for GSEs. And we also volunteer. I volunteer with the Western Fire Chiefs Association. I'm a member of the CAL FIRE Risk Modeling Advisory Workgroup that Chief Berlant has been talking about. And so my work is all around catastrophic risk, climate risk, and what makes insurance markets fall apart.

And how do you put them back together again? So I'm scrapping my fancy PowerPoint because a lot of the stuff that I was going to try to convince you, all of you seem to have already kind of glommed onto. And I'm going to try to talk about bigger picture things that you've been asking questions about. First of all, what makes a sustainable insurance market? Like some of the concepts that you've been asking about.

We have to have three components. The first is availability, which is what you're very focused on now, as you should be. And the necessary requirements for insurance companies to make their products available is they can manage and measure the risk, and they can charge premiums enough for that risk transfer. So if those two components are there, they're much more likely to be making their products available. Also affordability, which pretty self explanatory. The buyers can afford the premiums that the insurance companies want to charge, and they think that the premiums are worth it, as well, if they don't have to buy.

And then the last thing, which we hardly ever talk about, is reliability. And that is that the framework, the system, is stable and fair and predictable over the long term. And that might include judicial environment, it might include the regulatory regime. Also that the insurance company is around for the long term. And that is dependent on the ability to have enough reinsurance in a catastrophic event to be able to pay the claims.

So in my opinion, crisis is caused by a laser focus only on affordability and at the expense of availability and reliability. And very crude tools that are designed to control affordability, that have accelerated and in some cases caused market crisis to get worse. Happening at the same time that climate change has increased wildfire risk, and other things happening like inflation. So when you put those factors, like climate change and inflation and reinsurance market cycle that always happens out in the world, you put those against a rigid regulatory system.

It's like you've got your steering wheel locked straight ahead, you got your speed set on cruise control, and now you find yourself on the Pacific Coast Highway. What insurance company would agree to that? What owners of insurance companies or policyholders, who would say this is a good market to be in? That is why we have an availability crisis. So we have to think holistically, and we have to remember who owns this risk. Insurance companies don't own the risk. California owns the risk. That's one of the reasons that I have enjoyed and benefited so much from interfacing with the fire professionals here.

Is they're very, very interested in driving the risk down. In our case, pretending that the risk is lower than it is, all it does is it hurts reliability, and it hurts availability. It does not benefit affordability. The only way to get affordable rates is to drive the risk down. And I think everybody sort of understands that. But it means giving up some of our old tools, like that backwards looking formula.

I do have one slide I want you all to see if. You have any doubts about leaving that formula behind. It caused the affordability crisis. That's something that's hard to understand. It involves math. Sorry. So if you look at slide six. Right back in, if you think back to prior to 2017, the 1991 Oakland Hills fire was the most destructive event in state history. And that one we lost 2900 structures. So in 2015 and 2016, there was a rate hearing of a major insurance company that wanted a 6.9% rate increase.

In part because their wildfire risk was increasing. And the CDI and two of the consumer interveners used the backwards looking formula to argue that wildfire risk was not going up. In fact, sometimes they argued that the risk was going down, and they concluded that the company had to decrease their rates. This was right before, this was while we were in a five year drought. I think if you all lived here.

Then you remember how bad the drought was, all the trees dying off. And then right after that, we had the Tubbs fire in 2017, where there were 5600 structures lost. Camp Fire in 2018, close to 19,000 structures. And so the question is, is that a fair system? Does it make sense for that insurance company to have had to drop their rates in 2016? And then we have these things happen in 2017. So go to slide seven, and you'll see what that did with rates.

When you look at 20 years of data, and I just put together tiny numbers, but just maybe listen to the words. I put together data for three of the top 10 insurance companies, so you can't see who they are, but altogether, there's a pretty large market share. 20 years of data through 2016, the catastrophe load was about 11%. So for every dollar of non-catastrophe homeowners' losses, we would get another 11 cents of wildfire. And that translates to about a dollar 11 of premium. You add in the 2017 year, and the cat load doubles, it goes to 23%. You add in the 2018 year. Can you click? Click. Click. Okay, there we go.

Add in the 2018 year, and the cat load increased to 53%. So a five fold increase in the cat load by adding two years of data for three of the top insurance companies in the state. So in two years, that would have indicated a 38% statewide rate increase on homeowners. So when we have the 6.9% sort of implicit cap on what insurance companies can ask for without going to a rate hearing, that would take five filings to get there.

And it's hard for insurance companies to jack their rates up by 38% in two years. They don't want to do that. It's not good for their market, it's not good for agents, it's not good for consumers. This is what the backward looking forward formula allowed and actually dictated for rate adequacy in California. So no one should believe that the risk increased fivefold. The formula doesn't work to measure wildfire risk. Just one more slide. The next one shows like the green line is cat modeled rates.

The blue line is experience formula rates. We did a study based on, like, 10% of the risks in the state. If you wrote across the whole State of California and you use a catastrophe model, if you're growing in the WUI, if you're taking on more risk, what does the cat model do to your rate increase. That's that smooth green line on the left. If you're shrinking in the WUI, that's the green line on the right.

Your rates are slowly drifting down. With the CDI formula, it's that blue little horseshoe-like shape. That shape has nothing to do with increasing or decreasing risk in the WUI. It just responds to the experience. So the point is, the formula won't let insurance companies take on more risk. It incentivizes them to drop risks so they don't have to ask for rate increases. It penalizes them if they want to add more wildfire risk. They can't get more premium until after the losses happen. So why would anyone, right now, take policies out of the FAIR plan?

That's why this formula really has to be thought of as a key obstacle for availability in California. The other questions that people have been asking, is this the Holy Grail? Is it the silver bullet? The answer is no. It's necessary, not sufficient. The reinsurance cost in California is a real cost of doing business. California is the only state that does not allow for catastrophic reinsurance costs to be passed along through rates. We do allow it for earthquake. It did not break the earthquake market.

We allow cat models to be used for fire following earthquake and for earthquake insurance outside of the CEA. I don't know if any consumers have ever complained about that. I don't believe that I've ever heard that any cat model had all of their IP and their algorithms and their data published on the Internet for all their competitors to see.

Somehow California has managed to regulate earthquake insurance and fire following earthquake within a homeowner's policy without doing all the things that seem to keep coming up when it comes to wildfire catastrophe models. Other states have internal experts or state commissions, or they hire experts to review cat models. It's something our group does. I actually think California could benefit from joining up with other states that have similar problems.

There's a lot of states in the West. These models are not state specific. So these big companies like Verisk and RMS and CoreLogic that you've heard about today, they could put a model through. One Review for all states where wildfire is an issue. And the states could join up and hire experts to do a great job. And really kick the tires and really make sure that they're doing their due diligence and then protect the IP of the modelers, split the cost, and they'll have the information that they'll need in order to do their jobs of regulating to make sure that the models are suited for the purpose.

So I do think that our culture in California with respect to insurance has been viewed as kind of a win lose false dichotomy. If policyholders are going to win, insurance companies have to lose. And I think we really need to get away from thinking about that. Stop thinking just about driving premiums down, and start thinking about driving the risk down. Thank you.

Thank you. We'll see if we have any questions from Members. Pilar?

Sorry. I'm sorry. I know everybody's trying to get out here. I hear what you're saying, and I agree. But the problem is, how do people afford it? If you're talking about a 37% increase or a 56% increase, how do people afford that? That's one of the fears that I have in my mind, moving to this model, right?

And when you use the example of, it hasn't collapsed the earthquake market, but there's a lot of people who don't get earthquake insurance because it's unaffordable. So I don't feel like it's apples and apples comparison. So how do you suggest finding that sweet spot of insurance companies needing to be able to cover potentially what they're going to have to pay out? And we know they like to make a profit as well, versus what people can afford to protect their investments?

That's an issue in lots of places. California is just one of them. But I guess I'd say the first way to get affordability is competition. Insurance companies, I mean, there's such a culture of not colluding and not cooperating. They have very, very different goals and different views of risk. And if you have a bunch of different companies who want to quote your policy, there will be some that are lower than others. And so the numbers that I showed on the slides were not what the cat models would do. What the formula in California did. I think in some cases, insurance companies didn't need that much rate.

But they should have been able to increase rates over time as the risk went up. So I think on a macro level, competition definitely helps. Predictability helps. Knowing what kind of a house you can afford and where can you afford it, that makes a very big difference with respect to affordability. But the biggest issue is mitigation. I mean, one of the things about. I'm sorry, I'm looking at you turning my back on the most important people in the room right now for me. These guys have taught me so much about how mitigation works, and fire can really be mitigated.

We can actually drive the risk down. And don't just think about insurance premiums and affordability, which does matter. But think about how many lives will be saved. Think about how many communities won't be burned down. Think about how many fires won't be burning at the same time where Cal Fires spread all over the place and can't do as good of a job to preserve the fires, to preserve the homes. Those are the things that will happen if we have the information to mitigate. And these two guys and I and others, including Mike Peterson from CDI, who's been part of our dialogue from the start, we've been focused very hard on how to incentivize mitigation at scale. Not just home level, but community level, and how to get the words and the data and the concepts out there as fast as possible and have them adopted by the insurance industry.

The insurance industry wants to write in California. They want to be able to tell the sheep from the goats. Right now they can't. They can't tell the difference between a very hardened community and one that is just doing what they always did. And they don't like that. That hurts in terms of risk management and risk measurement. That's true for reinsurers, too. We've done a number of discussions with reinsurers and insurers that are big in California and said, this is what we think we should do. Will this help?

Do you think this is a thing? Will you join us? Will you relax your normal posture of never cooperating with anybody and try to get together on this data initiative and try to help get cat models to be more able to see what's going. On, to drive down the cost of inspections, make them more frequent, make them aggregatable within a community. And nobody has said no to us. Maybe because Frank and Dave Winnacker and the CAL FIRE connections make it hard to say no, but they're on board so far. I think it actually is good news. I think it's hope.

May I add. Would it be good to know that you've actually already solved this problem once? You've solved it for many years? How many of you are familiar with the Unfunded Accrued Liability in PERS?

Well, essentially we contracted. There's 3000 roughly agencies, 2 million people, I think, covered in it. And we realized some time ago we started to have a disconnect between the understanding of the PERS compensable items that were in thousands of different labor contracts and what was really PERS compensable out of the system. And we discovered that through actuarial science, by the way. We realized, oh, we've got a disconnect between the financial risk that we took on and the actual financial risk that we're responsible for.

Now, the bad news about that one is that's financially expensive. We know that if you're an administrator, you write a UAL contribution every year in addition to irregular. I believe we have an unfunded, accrued wildfire risk liability. Climate change, accumulation of fuels, and development into the WUI have been the incremental changes over time. Historical data doesn't work anymore because it's outside of the current conditions on the ground. And so now that we really understand the risk, there's two things.

One is we have to properly price that risk into the future. And by the way, if we don't, what starts happening to mortgage industries or municipal bond ratings? If the true risk is understood, we have a huge gap. I'll let your imaginations take that wherever they go. The good news is, to your point, we don't have to directly throw dollars at a lot of this. We should. There's a lot of areas that we are and we'll need to continue.

But if we could get to Chief Berlant's point, an aligned message out there to homeowners and communities on what's actually effective and what's going to be recognized in a material way meaningful to them, that's where this worse has to get done at scale. And so the way we buy down that risk, that unfunded accrued welfare risk liability, is through mitigation. In some cases that could be as simple as an afternoon with a good drink and a rake. And in other cases, it's going to require grant support because you've got 100 foot tree between two or three structures that are 20ft apart.

But with better data and alignment behind the message and a strategic system, which is addressed in my written comments, we can now take this kind of policy fibrillation that we have and put that into an effective mitigation heartbeat. There's a way to do this smarter, but it requires us to work in a transdisciplinary approach. And so you've got, in the background, you've got a real cohort out there sort of unofficially working in some other areas, and it connects here and it connects there. But we're close to starting to hit critical mass on that.

Cat models is a step in the right direction, I believe. And we need to do these other things to help that be more granular, more accurate, and send a consistent message to all the stakeholders. Four groups have to be in alignment. The State Insurance Commissioner Guidance, the insurance trades in each state. I'm doing this work in California, Oregon, Washington, Nevada, and Colorado. In each state, we're taking the same approach. It has to be the state organization responsible for WUI because that sets true north for all of the other agencies, counties, cities, special districts.

It sets true north for them. But you have to get the local fire agencies to complete that last mile. And so those are the four groups that need a consistent alignment around core mitigations at the parcel level and the property level that are evidence-based from a fire perspective, which means they're now evidence-based on an actuarially financial base because they work. So help us follow the science. Thank you.

Thank you. Assemblywoman Waldron, do you have a question?

Thank you, Madam Chair. So as one who lives in a very high fire risk area and has been on the fair plan a couple of times and has been non-renewed so many times that I can't remember in the different homes that I've had. I know I always said the fair plan wasn't fair because we went from, I mean, talk about 10 to 12 years ago, it only cost $1,000 to insure the house, and on the fair plan it went to $7,000.

So we're talking about huge, high change in the premium cost, and it doesn't cover near what normal insurance covers. And then other times, and right now I'm lucky, I got actually a policy from State Farm which now I know will not be renewed at the end of the year. My neighbor is on the fair plan paying $24,000 a year.

Wow.

I mean, you're talking about the life investments of people that, if we don't solve this problem, we're going to see an economic crash like we've never seen before because people are not going to be able to afford their homes. Coming from San Diego, the area that I represent, we saw the Cedar fire in 2003. I was on the City Council at that time. We had 2820 structures that burned and we lost 15 people. In 2007, we had Witch fire, which was another huge fire, almost 200,000 acres.

I think it was like 197,000 plus acres. We learned a lot since then in San Diego. So we haven't, keep our fingers crossed, seen that type of catastrophic wildfires. We've had fires with lilac and others, but we learned and SDG&E created its weather center, which is really top of the line as far as being able to predict the chances of a fire where it's going to burn. It takes the historic data, but it takes all the other conditions into effect, and the winds and humidity and all our fire chiefs have access to it on an app on their phone.

So as soon as something happens, we can shut down in San Diego, 10 homes instead of 100,000 homes, because they know exactly where this path of the fire is likely to go. And it helps with evacuation as well, because then the fire chief can look at it and say, okay, with these limited backcountry roads and in and out, one way in and one way out, we can't have the whole community trying to get out on the road because then you have people dying and they can't get out.

So they can look and see exactly where they need to evacuate, just these homes and not all of this. So all of that data is available and has been available, at least in Southern California. I think Northern California still has a ways to go as far as that. I being frustrated, having been on the fair plan at the time, I wrote two pieces of legislation that had to do with almost exactly what we're talking about. And having the insurance companies who are using modeling and what kind of modeling, have the insurance Commissioner look at those things.

Also taking into account, the other one was if the insurance company is using these modelings and they're not just zip codes where you just blanket non-renewed people from every zip code without looking at individual homes or risks, that that would help the individual homeowner keep their home. If their home hasn't burned in 40 years, then why is their risk as high as 3 miles down the way that the fire has historically, every 20 years or so, burned through? But these bills never saw the light of day.

And the reason is because the insurers, and I'm looking at you because I'm wanting a response, at that time in 2019 and 2020, said we can't do that. We can't be more specific with the communities or with individual homes because we need to spread the risk in order to keep solvent and in order to pay. So what we're talking about, has that changed? What's the difference between 2019 and 2020 and now? Because I'm looking at if we can use this type of data, I'm hoping more insurance can enter the market, and I'm also hoping that prices will go down, or at least be within reason.

Sure. Sure. I'm sure there's some lobbyists in the audience that would just be jumping up and down to answer your question. They're going, Nancy, please don't screw it up. They'll be bitterly disappointed. So I would say it's impossible to expect one insurance company to write a community. And a bad idea, honestly. You don't want that. You don't want them to have that kind of concentration of risk. The data that you're talking about, they'll need to be doing inspections every year in order to continue to understand the vegetation around each one of those houses.

Some companies charge $150, something like that, to come out and do an onsite inspection. That's not really sustainable. That's a lot of money, really. And someone who inspects my house does not know anything about my neighbor's house. That's what the stuff that Chief Frievalt was talking about. We're trying to put together a collaborative data coordination campaign which would allow one company to use data that another company got from an inspection from the year before, and maybe allow fire-wise communities to do their own data collection.

It doesn't even have to happen from the insurance industry. Let Cal Fires data go in. Let protected fuel brakes go in as another data point for modelers. If they can rely on having data across the state, you're talking about your area which is great, anecdotal and spotty data can't be used operationally by modelers or insurance companies. It has to be available at scale. It has to be auditable. It has to be standardized, or else. That's where I come in, like, how does the process take this in at scale, and how do we recognize your community?

We know that Cal Fire thinks you're in a very high hazard area, perhaps. But what are the three things that you did to drop your community risk down three levels? If you classify communities that way, insurance companies will use those classifications in underwriting and rate-making. What they will do with it, I cannot promise, but they will use it if they believe that it really makes a difference in discriminating sheep from goats. I want to mention one other hopeful thing which really gets to affordability for existing communities.

It's possible that building new housing outside of a risky community could reduce the risk in the community if you even. You want to do other things, but where I live in Orinda, we've lost a lot of insurance companies as well. And I think if new housing were built on the outskirts of the town and it reduced risk inside the town, wouldn't our town feel a little bit differently about building condos and affordable housing that doesn't exist right now? I'm not promising that will happen, but that's an avenue that I think could be explored. That should be a win-win for California.

Well, do you have another question?

I'm just thinking, you know we talk about mitigation and everything. On my six acres that I have, we clear the weeds 4 and 5 times a year. So you could inspect it and then three months later get the light winds or the rains. All that does is water the weeds, not the grove, so there's the weeds again. It's like a changing situation.

We have to have a process that takes that into consideration and makes the most with the least amount of work and the least amount of cost and that requires collaboration.

Yeah, I mean, I'll just say this one thing. We have this one canyon where the wind sometimes can increase up to 120 miles an hour during a Santa Ana. And with the right humidity levels and everything, the fire chief there in Rincon calls that the mouth of the dragon. And if we could control that, that would be a mitigation. I could see that would help, maybe.

That's where the fire pathways that Chief Frievalt was talking about, we're working on something with the town of Paradise. We just got a brick grant this week and we're working on looking at fire pathways into Paradise and looking at ways to disrupt. It's not a peanut butter spread of mitigation. This helps affordability too, like targeted dollars.

We might find that rich people on the outside of town are the ones where you put all the money in and the houses maybe by less wealthy people in the middle of town, might benefit. So it's a sort of a weird thing where you have to figure out, where do the dollars go to drive the risk down for everybody?

Sorry, this is the last thing. Because we had an effort to get 25 million for apparatus for fire districts in high fire risk areas to clear backcountry roads just along the roadways, because in San Diego County, the data shows that about 25% of those wildfires start from a vehicle that pulls over to the side of the road.

Wow.

But we didn't get the money, so that would be another effort I think we could do statewide.

Well, thank you so much for being here today. We appreciate all your comments and your patience and answering our questions. Now we're going to open up the microphone to public comment. Thank you. Please come forward. State your name and affiliation and please limit your testimony to two minutes each.

Dan Dunmoyer with the, excuse me. Dan Dunmoyer with the California Building Industry Association. Chairs, thank you for holding this really important hearing. Just a couple things to build on from all the comments. We do support catastrophe modeling. We are the only building industry association in America that has supported the most strict fire-hardening standards of any building industry in the United States. Since 2010, here's a positive note on today's hearing.

Since 2010, since we've been using the newest codes, not a single one of our master plan communities with regular mitigation have burned. So when you look at all the horrible fires we've seen, we have created a fire science for building in California that stops wildfires from destroying our communities. I was stunned by Nancy Watkins' comments. We do believe, and two major fire chiefs have said, if we were to build rings of new homes around old homes, we could solve this crisis and solve the housing crisis.

Two other key points we do support catastrophe modeling, particularly if it includes recognizing mitigation. So one of the biggest arguments to the insurance industry is, hey, you cleared the field today, that the Assemblymember mentioned, but a year later there's high brush. We believe now that we have master plan communities with homeowners associations, we build in regular maintenance of fire-hardening. That needs to be taken into account in the modeling. It's not quite there yet. It's not accurate.

We have Google Maps, we have other sophisticated tools, but mitigation has to be recognized. As was recognized earlier, if you have a community that doesn't burn because of mitigation, that needs to be taken into account in the modeling. Last thing I'll say is we are sitting on a housing crisis today and it's getting worse in this fire situation for one key reason.

The Commissioner has done some really good work in expanding the fair plan to 20 million per location, but most condo associations are bigger than that. Last night I was told in San Diego County that a project to build 1100 condominiums is now put on hold because there's no adequate commercial coverage in the fair plan or anywhere. Last thing I would say is as bad as 38% rate increase is, it's a bad number, our record this year, one of our members went to place insurance after their Association was non-renewed.

The difference between an admitted carrier and the non-admitted carrier was 6900%. Not 38%, but 6900% went from 40,000 to 2.8 million. For the monthly association, dues went from 250 to 1500. So as bad as a 38% rate increase is, it's profound, we believe that a robust market will result in a solution to this problem, and we're committed to work with the Chairs and the Members to do this.

Last thing I'd say is we need to act quickly. This is real. This is affecting homeowners today as we sit here, and we'd like to be part of the solution. Thank you.

Thank you.

Madam Chair and Members. John Norwood on behalf of the Independent Insurance Agents and Brokers Association of California. I'll be very quick. Our members tell us, as we've said before, this is the worst market they've ever seen in their careers. We get regular reports from our members that have a client that can't expand their business because they can't get insurance. They're facing foreclosure on their home because they, or a high force place policy, because they can't get insurance that are worried about losing their condominium because they can't afford the homeowners dues increase to pay for the insurance.

We started this conversation on rate filings and reinsurance and things over two years ago, and it's a snail's pace. I agree with Mr. Dunmoyer, something's got to be done. Yesterday on this, I did testify one of the last oversight hearings you had and indicated that everything we've seen and the work that we've done that we've hired out is most companies are looking for a 20% to 40% rate increase to get to adequacy, and most of those won't apply that overnight. But that's what they need to get to rate adequacy.

So if you take an average 30% increase on a typical homeowner's policy of $1000 to $1,500, yes, it's three hundred dollars or four, but it's a couple of dinners out or something. It's manageable, is what I'm trying to say. For an HOA that's paying 50 or $80,000 a year for homeowners insurance, a 30% increase is manageable. 1000% or 15% or 500% isn't manageable. So again, I agree with Mr. Dunmoyer that we need to move quickly, and I appreciate both Chairs and the Committee staff for the work they've done in putting this very good hearing together. Thank you very much.

Thank you.

Yes. Yes. Thank you. Good morning, or afternoon almost at this point. Seren Taylor on behalf of the Personal Insurance Federation of California. I want to thank you and all the panelists for the thoughtful discussion today. We did submit a detailed comment letter that examines the many challenges facing California's insurance industry, which is noted in the excellent committee background or extend beyond the one factor examined today.

And to Assemblymember Schiavo's point, it's not a panacea, but it is a very important component of an overall comprehensive plan to get a more resilient insurance market back in California. I think Ms. Watkins' testimony provided a great overview of the need for forward looking catastrophe models instead of the current backwards looking approach that fails to recognize the impact that climate change poses to California properties, and it actually discourages insurers from increasing availability of coverage in wildfire areas.

And the simple fact is, if California law won't permit insurers to develop rates using scientific understanding, then it's likely insurers will continue to limit issuance of policies in high risk areas. And we don't want that to happen, right? These companies have spent decades building brand. Their agents that Mr. Norwood talked about have their clients. They care about this. They're not in the business to non-renew people. They don't want to be making these decisions.

We have a system that is leading to a predictable outcome, and there are other regulatory and process issues that we look forward to working on with our colleagues at the Department of Insurance and, of course, with the Legislature and the Governor as we move forward. I do think this hearing has clearly demonstrated that catastrophe models are an extremely valuable tool for both insurers and policyholders. They provide policyholders also with insights into risk mitigation.

By understanding the vulnerabilities associated with their property or location, they can take proactive measures to reduce their exposure. And that's consistent with the new CDI mitigation wildfire risk regulations. So, I know there's more work to be done. We really look forward to working with all of you on that and appreciate the opportunity to provide these comments. But we really need to work on a comprehensive solution that protects choice and insurance availability for all Californians. So, thank you.

Mark Sektnan with the American Property Casualty Insurance Association. At the request of staff, I've jettisoned my 20 minutes of public comments. I just want to say thank you for the hearing. I think we learned a lot about the role that catastrophe models can play. Catastrophe models have come a long way since the 1980s when everybody talked about these black boxes that nobody knew anything about.

In addition to all the verification tools that were mentioned earlier, the National Association of Insurance Commissioners is working on a center of excellence to help review these models, particularly for small states that don't have the staff that California enjoys. I think the message is very clear and has been said in other hearings. We can't insure ourselves out of this mess. We have to mitigate. The only way you decrease rates is by decreasing risk.

I want to say that we appreciate what the Department of Insurance has been doing. I've been doing a lot of work with the Department of Insurance to try and facilitate conversations, to try and bring insurers back to California. They want to write in California. They're watching very closely, and they're hoping that we're able to accomplish something here. Thank you.

Thank you. Okay. Thank you all for participating today. This Committee is adjourned.