Resources

• First Street: Climate risk analytics platform connecting climate science to financial outcomes.
• FEMA Flood Maps: Traditional flood risk mapping tools referenced in the episode.
• NOAA / NASA Climate Data: Government sources for climate science and hazard data.
• Munich Re Reports: Insurance industry reports tracking climate-related losses globally.

Host and Guest

Host: Mike Toffel, Professor, Harvard Business School (LinkedIn)

Guests:

John Mulliken, Senior Lecturer, Harvard Business School (LinkedIn)

Ed Kearns, Chief Sciences Officer, First Street (LinkedIn)

Transcript

Editor's Note: The following was prepared by a machine algorithm, and may not perfectly reflect the audio file of the interview.

Mike Toffel:

Ed, thank you so much for joining us here on Climate Rising.

Ed Kearns:

Yeah, thanks for having me.

Mike Toffel:

So why don't we begin, with a very brief introduction to your role and a bit about how you got there.

Ed Kearns:

I'm a scientist. I'm an oceanographer by training. And I work at a place called First Street where we are connecting climate change to financial outcomes. And I got here by a long career in science, both at the academic level at University of Miami and also with the federal government, particularly with the National Oceanic and Atmospheric Administration. We're dealing with a lot with climate data and climate outcomes. And I was inspired by the difficulty in explaining climate change impacts to the American public and to American businesses to try to find a different way of doing that. And that's why First Street exists. And so I joined First Street about six years ago after 15 years with the federal government to work on climate.

Mike Toffel:

Great. And so let's talk a bit about First Street. So what does First Street do? Who are its customers? At a very high level, we're going to dive in for sure during the conversation.

Ed Kearns:

Yeah. So we translate how climate change is impacting physical risk. So this could be flood, this could be wildfires, this could be winds, translating that into financial outcomes. So if people need to be making decisions either for themselves and their home or perhaps the business that they run, then we provide the tools with which they can understand the climate science, translate that risk into some kind of financial measure that they can then use to make decisions about how they're going to be impacted.

Mike Toffel:

Got it. And is this focusing primarily or exclusively on real estate, or are you also thinking about supply chain and agricultural fields and things like that when you're thinking about these risks?

Ed Kearns:

Yeah. So we're looking at physical risk as it impacts buildings and properties right now. This is starting to also go into infrastructure and also supply chains. We have stopped short on things like agriculture, which already has a very active industry looking at some of these things. So there are certain pieces of it that we haven't gotten too deep into yet, but we are describing the hazards for everywhere on earth. And so they can be translated into these other things as well. But right now, we have seemed to have found a good product market fit with real estate and with investors in real estate, and that's where we're going right now.

Mike Toffel:

Great. So John, you want to dig into the science of physical risk?

John Mulliken:

That's great. Yeah. Ed, I'm really interested in how you compare to some of the existing models. I mean, there's traditional risk tools. I think a lot of people are familiar with FEMA flood maps and they were really built for a stable climate. And so I'm just wondering, how do you design a model that's explicitly forward looking, one that captures what one of these assets you're looking at is going to face in 15, 30 years, not just what it faced historically?

Ed Kearns:

Yeah. And the FEMA flood maps are a great example of the challenge in front of us. So FEMA flood maps were created, like you said, for a stationary climate. They are looking only at riverine flooding and surge flooding. They don't take in account heavy rainfall, for example. And so, but what we've discovered as a country over the last 30, 40 years is that a very large percentage, about 40% of flood claims are coming from outside FEMA flood zones in the US. It's been very consistent over the last couple of decades. And the question is, well, why? What are we missing? What's missing from this equation? And it's been largely that the lack of heavy rainfall flooding as included as part of the FEMA flood mapping, if you don't include rainfall, you're missing a big piece of the puzzle. It also happens to be that piece of the puzzle that's changing the most with climate change right now.

Because of the relationship between air temperature and water vapor in the air. For every one degree centigrade that you increase air temperature, you can cram in 7% more water vapor. And so if you're living on the east coast of the US and you think that rainfalls have gotten heavier over the last couple of decades, you are not imagining that. It is demonstrable that that is what's happening, right? And so at First Street, what we're doing is we're relying on the physics-based models. So for flood, we're going to drive a flood model that shows how water will move over their surface and we're going to drive that with a rainfall and we're going to drive it with river and flooding and we're going to drive with surge flooding and we're going to drive it with sea level rise, bring all these things together and say, "Well, just how deep can the water get at your business? How deep can the water get at your home or at that infrastructure? What's that look like?"

And then we can do it probabilistically, look at different, what's called in business return periods, how often you would see this kind of storm come back or this kind of configuration come back. And you may hear of the hundred year storm or the 500 year storm. So those equate to a 1% annual risk or 0.2% annual risk. So you could do this probabilistically in rolling the dice and see what are the chances of having water that's going to come into your home this year and how that is going to change in 30 years or a hundred years with climate change. And so by using these physics-based models, we can map out those probabilities to a great degree of accuracy and arm you with that kind of information that you need to make a decision.

John Mulliken:

And I imagine that there's pretty deep uncertainties still in these climate models that you have to face scenarios in which things can behave non-linearly, in which you can see that the trajectory is going to change over time and perhaps in ways that might be hard to predict. How do you handle those ways in which the climate system begins to move in ways that it has historically moved in response to what we've put in the atmosphere?

Ed Kearns:

Yeah. So we are focusing on those things for which we do have predictive capability. So that is air temperature in particular, right? We know how that's changing. Sea level rise, another good one. And by looking at the studies of hurricanes and extra tropical cyclones and stuff, we can see in the climate models, and there are dozens and dozens of climate models that have been produced, we can use groups of them. So we're basically doing ensemble modeling, right? We're taking the averages across multiple dozens of climate models to drive down any one model's uncertainties. And then we're looking over a span of 20 years. So when we look ahead 30 years, we're taking plus or minus 10 years around that 30 year target. So again, we're averaging many years together, again, driving down the uncertainties.

John Mulliken:

That's great. I was wondering really about, you've mentioned a number of different types of climate risk. You've got fire and you've got flood and so forth. And you mentioned how as temperature goes up, the atmosphere can hold more moisture. And I'm interested in how these different types of risk interact. Do they compound? Are they additive? How do asset owners think about the combination of all these different risks?

Ed Kearns:

Yeah. They're actually in phase with each other, right? There's so many different oscillations within the climate system. Your listeners may have heard of something like El Nino, right? They may not exactly know what El Nino is, but this is a type of climate oscillation. There's literally dozens of them, the North Atlantic oscillation or whichever, pick your favorite oscillation. But what that does is it ties climate events and climate risks together on continental, even global scales. And so when you're modeling climate, it's one of the challenges, right, is you have to consider it as a global system. And all these climate models I was just mentioning, that is how they're simulating climate. They're looking at the global expanse. It's very difficult and very expensive to do, but that is the current state of the art in modeling. So by looking at these things, not as isolated events, but looking at them together, yes, sometimes they do compound.

And so what we've done at First Street is beyond modeling the risk from that particular hazards, say flood and wind, is then we look at the correlation between those perils, as we've seen in observations over the last 20 or 30 years. And that is something that you, again, can measure and you can put a number on it, right? And then you can map that correlation into future estimates of losses. So if you know that these areas are under certain climate conditions are always beating together or maybe they're beating apart, maybe one thing might make another type of risk more likely or make it less likely. So you may be compounding losses and risk or you may be diffusing that. But all those evidence for that is in the observed record and we have very good observations of these things over the last couple of decades. So we lean heavily on the observations of the past and we're mapping that into the future with these correlations already in place.

Mike Toffel:

So Ed, I wanted to ask a question based on what you said a moment ago, which was that you have this ensemble model, which is sort of a model of models. And is that the IP, the intellectual property that First Street offers? Because you're not developing first order models, you're relying on other models, but what you're doing is assembling them. Is that what you're doing or are you also offering yet more new models or tweaked models? Because you were talking about comparing predictions to actuals, which makes me think about machine learning and now you're coding it and maybe improving the models. So which are you doing? Or maybe some of each.

Ed Kearns:

It's some of each. So the averaging across models is nothing particularly novel. That's a very common method. Your listeners may, if they live on the East Coast or Gulf Coast of the US, I'm sure they've looked at hurricane forecasts before and something that the National Hurricane Center does is that they use ensemble modeling also too, and they may have seen what we call spaghetti diagrams in the business, right? Lots of different models and the forecasters then will say, "Okay, well, you've got 15 different models that say the hurricane's going to go here. We're going to average them together and we're going to have our cone of uncertainty." And so that has been a very effective approach, right? Not just figuring out where the hurricane's going to go, but communicating that to the American public. So kudos to the Hurricane Center, kudos to the National Weather Service and NOAA that have... And they continue to improve this, right? So it's a very effective way of moving this forward.

Now, we're faced with a challenge of taking these climate models and making them useful for individuals. And so to do that, as I said, the climate models are being run on a global resolution, and necessarily, because of the computation and cost involved in that, they are relatively coarse in resolution. So they're 100 kilometers, 50 kilometers. If you think of it like a pixel in a digital image, each pixel is a hundred kilometers wide. So if you're trying to figure out what's happening at your business or that piece of infrastructure or your home, you need somehow to take that information and we call it downscaling, getting down to the finer level. And so, and this is where First Street is adding value, and this is where all the work is, how do you get down to that 30 meter level in wildfire? How do you get down to that three meter level in flood? Because it matters whether the creeks on one side of your house or the other, right? It really matters.

And so we do that by injecting a flood model or a wildfire model or a wind model between the climate models and then the target, which is an individual building. So we have to have some way of translating that coarser information into a high resolution prediction that can be useful for people so that you have house to house or business to business differentiation if one building's a little bit higher than the other or something like that. So this is something that you don't see very often because it's hard. So in the theme of flood models that we were just talking a little bit about before, they are famous, or I should maybe say infamous for having maps drawn with a big border around an area that has flood risk and everything inside that is binary. You either have flood risk or you don't, right? And it doesn't matter if one of those homes is higher, maybe it's been raised higher than the other ones and it's going to be above the flood. They just consider them all together because it's simpler to model that way.

So what we've done is, yeah, we've spent the time and the effort to tune and run these physics-based models, not AI. We're using physics-based models because we need to be able to trust that downscaling. We have to understand the physical processes that are allowing that downscaling. Not only that, we have to also be able to trust that 30 years into the future, that our projections 30 years are also going to be trustworthy, right? And so the physics is not changing in 30 years or 100 years. Physics is going to be the same, right

John Mulliken:

And you're not the only company that's doing this. And there's been this debate that I've read about that your damage estimates might in some cases run higher than other ones that are out there. I mean, as a scientist, how do you think about that? Is it better to be more conservative or better to be more aggressive? How do you make the case for accuracy when you're projecting these 15, 30 year risks and you can't perfectly validate them yet?

Ed Kearns:

Yeah. So it's really, a lot of the differences among the models is what they were created for. So some of these comparisons, and it's great to see the comparisons. There's not enough of them out there, quite frankly, so we need to see more of these done. But there was a paper recently looking at some of the insurance industry's CAT modeling comparisons with First Streets'. Those models are all built for different purposes. Some are built for the insurance industry. They're built to be calibrated on the losses themselves, and they're looking at extreme events. They're only looking at the hundred or 500 year events. Now First Street, we're looking at risks all the way from, on the coastal areas, the one in two year, the every other year type of thing. So we can resolve things like king tides and stuff like that, right? And then all the way to one in five year, one in 20 year.

So when you're adding up losses, those events with the lower return period are not as severe, so they don't do as much flooding, but there's a lot more properties that have that risk and it's more likely to occur. So if you're creating an annual average loss product, which we have done, and you start to include those lower return periods, you will get a lot of losses. They're small losses, they're not extreme, and they're not the kind of losses that the insurance companies are necessarily worried about, because these aren't the things that are going to break the bank for them, right? But these are the kind of losses that are going to be impacting individual homeowner or the individual business owner. And so they need to be aware of them also too.

So we're building the model for a slightly different purpose. And so that's, I think, one of the reasons why we're seeing some of the disparity in some of the loss estimates, just because we have a different target. And so if you start to slice the data more finely and looking at only at one in 100 year risk or something like that, you may see some other differences because the inputs are a little bit different also too.

John Mulliken:

Is there an incentive? Do clients want to see more risk or do they want to see less risk?

Ed Kearns:

No, that's not what I've really perceived in the business. People just want to know really, what is the risk? What can I bank on? And this is a good sign in the community that with climate risk, it's being now looked at not as some esoteric thing, but something as like, "Oh, this is just another risk that I have to manage for my portfolio or for my business or for my bank." It's just something else and they're managing lots of other kinds of risks. Now climate change now is saying, "Okay, this is yet another one that I've got to manage." So in the other risks that they're managing, there's uncertainties with those. With climate risk, with the physics of the climate system are actually pretty well known. And so majority of the uncertainties actually don't come through this physics-based modeling right now, but they're coming through the damage functions and the loss functions associated with the physical damage and downtime to buildings. There's a lot of variety in that depending upon the types of buildings.

And so you're seeing a whole resilience industry also too now stepping up to the plate, which is great to see. The engineers are coming out and saying, "Hey, yes, we can help people mitigate for this risk." Because I think the other thing that's really happened in the last two years is that the talk in the industry has turned away from, is this a risk? Is there transition risk to, "Oh yeah, we've got this risk. It's going to happen. It's a sure thing. How do we mitigate it?" And I think that's a very... It's maturing. It's becoming a healthier conversation now.

Mike Toffel:

Yeah. I mean, when people talk about transition risk, typically they're talking about policy or norms changing to price carbon. How does that change the market for renewable energy or for EVs? In this case, what we're talking about physical risk, that's happening regardless of policy. And in the long term, the question is just depending on how much we can mitigate our carbon emissions that will mitigate the amount of physical risk we will face, but we're facing physical risk in some manner no matter what. So it's a really different conversation.

Ed Kearns:

It is. It's a sure thing. And this past year in New York during Climate Week, during the UN General Assemblies, it's called Climate Week, a lot of climate people come to New York to have these discussions. And in previous years, I've often had to explain what you just were talking about, that the physical climate risk is a sure thing because everybody's talking about the policy and the transition risk and curbing carbon and how can we change the course of all this. And I would have to explain to people that, "Well, even if you institute a lot of these great reforms that are going to have impact on climate, that we're not going to see the impact of those things for 20 or 30 years because the train has already left the station." This year at New York Climate Week, I did not have to explain that not a single time to anybody. So everybody now is like, "Oh yes, it is happening. It has happened. I must prepare." So like I said, I think it's maturing.

Mike Toffel:

Great. Well, this conversation about models and the science of physical risk, that's really about the product or services that you offer. Let's talk about the customers who are interested in that product. And I think there's a number of them, right? But you tell me, I can imagine real estate investors thinking about assembling their portfolio. There's real estate developers who are thinking about a particular site or there's companies who are thinking about putting a factory, a billion dollar wafer fab. And then there's the consumer side thinking about buying a home, which constitutes a large portion or vast majority in some cases of their savings and for retirement and so on. So take us through a couple of use cases for different customers and how you're engaging with these customers.

Ed Kearns:

Yeah. And I'll do that by maybe also explaining at the same time a little bit how First Street evolved over the last six years also too, because when we started releasing our first flood model back in 2020, we were a nonprofit at the time, and we were really trying to figure out how to communicate this risk to the average everyday person, because we weren't sure yet what that market looked like, but we knew we had to figure out how to create these products and how to communicate them. And so yeah, some of the first things we had done with flood was publishing maps on our own website to say, "Well, this is how it's changing." Doing things like simple numerical scores, one to 10 scores, one being good, 10 being bad to simplify the presentation of that information, just like I was describing earlier with the Hurricane Center, how they've managed to communicate uncertainty and risk in a certain way.

We're trying that same thing with flood risk. And I think we discovered that, yeah, there's a big market for that out there. A lot of people want to know about it. A lot of homeowners are concerned about this because they think that they don't have a good view of risk from FEMA or from their local authorities. They're not quite sure what it is. And so they want, number one, to find out what their risk is. If they find out if they do have risk, then they immediately go to, "Okay, how's this going to impact the value of my home?" And so for most Americans, that's their biggest investment that they've got, is the money in their home, and they want to learn how to protect it. How do they get insurance? How much insurance should they have? All these kinds of things. So we did experience that, and then soon thereafter, we discovered certain US federal agencies, whether it was Treasury or the Federal Reserve banking system, Fannie and Freddie, FHFA, they were all very interested in this too because they had the same questions, right?

But they were thinking on a large scale, thinking national scale and how much climate risk is inherent in this. And I was also concerned about that. I grew up in Miami and left Miami in 2008 during the big 2008 downturn. I don't want to see that again. I don't want to go through the real estate issues that we had back then. And so anytime you have a latent risk that's not well quantified like we had back in the day, there's a risk that this could get bad, right? And so the Office of Financial Research at Treasury had put out a couple of reports, one that looked at Freddie Mac's portfolio with risk. And this came out like in 2023, I believe, and they had analyzed independently how much climate risk was in Freddie's portfolio, which of course is on the backs of the taxpayer. And they discerned that no, the climate risk was being managed well by Freddie.

And so this was music to my ears. It's like, okay, great. This information that we're putting out there is being used to manage portfolios of real estate and mortgages in a very clever manner so that the risk is diffused and it's not going to spike on us. It's not going to really hurt the housing market like we saw in 2008. So this is great. Then banks also started to follow along in suit and also worried about their mortgage portfolios. And there was regulation at the time that was also, they were interested in finding out how they're going to meet some of those regulatory requirements that they were facing at the time. And so we also started talking with banks.

And then of course, with the new administration, there were a lot of changes of priorities and a lot of that climate regulatory work was de-emphasized or removed. And so the banks weren't as interested anymore. They didn't feel the urgency anymore, but real estate investors and asset owners, asset managers, now we're starting to step up because this information was becoming more widely known and they realized, oh, this is a material risk to our portfolio. And maybe it's a pension fund or somebody that's really, they're having a longer outlook, they have hold periods that are beyond eight or 10 or 15 years, they are the most concerned about this. And so, oh, here's a market for this kind of information, but you also have to be able to provide this for thousands of properties and you're not going one by one, you're doing it portfolio by portfolio.

And so what we've done at First Street is we've continued to, as the market has changed and merged as we've learned all this, is we've brought more tools to the table. So initially we were just creating basically mounds of data that if you're an analyst at the Federal Reserve Bank, they're fantastic analysts there. They could handle us just dumping our entire database on them. They can analyze it, they can work their way through it. But if you're a smaller pension fund or something, you don't have that kind of horsepower on staff that can handle that. And so what we had to create at First Street are the software tools to slice and dice the data, organize by portfolio, do the CAT modeling, do all the projections necessary to give a very refined sense of what that risk is. So it's a useful tool for them as they're making their decisions because they're making these decisions every day.

John Mulliken:

That's great. Can you bring it to life for us? I mean, if I'm a client of First Street, what am I looking at when I ... If I have a multi-tenant housing development that I'm looking at understanding the risk for, how do I interact with First Street's products?

Ed Kearns:

Yeah. So if it's an individual asset like that, you'd be looking at a map of where that asset is and you'd be able to see the spatial area around it and where the flood or the wind or the fire risk is. So we show the pattern of what the hazard is around that property. And then you'd be also then looking at an analysis of what the intensity of each hazard is for that property. So if it's flood, for example, maybe the projection is there'll be two feet of flood there in a one in a hundred year event. Then you would say, "Okay, if there's two feet of flood there, is it getting in the building or not? How high is the building?" Maybe the building's already been raised up on piles or it's got some kind of retaining wall around it so two feet doesn't concern you.

Or maybe you say, "Well," or maybe it's 20 feet and it does concern you and, "I have to do something about this. I have to get some insurance or I have to do something to mitigate this physically at the site." It'd give you the information you need to do that. We would also give an estimate of what that mitigation cost would be so you could do that return on investment. Is this going to be cost effective for you to put this in, if you're going to hold it for 10 years, will you get your money back? So we provide that kind of level of granularity in the decision making too.

So if it's an individual property, we try to describe everything about that property, all the characteristics that we know about it. And of course, sometimes either the public records about that property may be incorrect or the satellite image of the property may not have captured something correctly. So we give the user also the opportunity to go in and change those parameters about the building. Maybe it's higher than we thought it was. It's elevated more or the roof type is different for wildfire risk or something like that. So you can go in and you can change those things too to make it correct and then see what that does for your risk for that particular building. And then you can make your decisions about what you're going to do with it.

Mike Toffel:

Got it. So that's for the individual. And then what about if you have a portfolio of thousands of properties?

Ed Kearns:

So you would take those thousands of properties, basically bring it in an Excel file or a CSV file and upload it to our site. And once you've uploaded to our site, then we'll analyze each single property the same way we did that one statistically. And you may choose to look at any one of those thousands individually if you like. But importantly, then we roll those up into an aggregate risk. And what's important there is an exceedance loss curve so that you can look at for this group of properties, what are the chances are? So you're looking at probabilistically looking into the future for a one in a hundred year event or one in 500 year event or one in 20 year event. What does that curve look like? What are the chances of you hitting a certain level of loss in that portfolio? So it allows you to say, "Hey, maybe that my risk is too high and there's an uncertainty bound around that also too." Maybe most of that risk that you're not prepared for is coming from only a handful of properties within your thousand property collection in your portfolio.

Maybe you decide to do something, divest yourself of those or mitigate those, do something there, invest some money to fix the risk there and watch your whole exceedance loss curve come down. And as long as it gets down to the point where it's manageable through either other risk transfer mechanisms or whichever you choose, then that gives you that information, that confidence that you can proceed with your portfolio and be sure that the investment that you're making is going to be secure in the years ahead.

Mike Toffel:

Got it. It seems like another key stakeholder would be a home owner or a home buyer at the individual level. And it seems like there's winners and losers all over the place here. If the risks are revealed to be less than was generally thought, well then you're maybe a winner. If the risks on the other hand are revealed to be higher than what was previously thought, then maybe you're a loser, if you're the home owner. For a home buyer, maybe you just want to know what is the risk. So if I want to invest a million dollars in this property, what's the risk profile look like? And if I have two properties that look otherwise similar except for these risks, well then maybe I'll buy the one that's less risky. What does that whole home buyer and home owner side look like?

Ed Kearns:

Yeah, it's been fascinating to see how this has evolved. When we first released those flood maps back in 2020, we were talking with a couple of the real estate sites, including realtor.com and redfin.com because they were interested in the same questions. And by studies that we've done at First Street, Dr. Jeremy Porter has led a lot of these econometric studies and socioeconomic studies, that yes, there is an impact on real estate prices.

Sometimes you have to tease it out because other market forces are causing real estate to rise and fall. But if you look carefully, you can discern that things like flood risk do have a disproportionate impact on how that property, how its value is going to change over time. And you have to look very closely. You can't just look at like county level aggregates and stuff like that. So I said I'm from Miami, Florida. If you look in Miami, and this has been going on for 20 years, right? The markets have already been at work even before First Street existed and had scores on homes. People were already moving to the high ground. They were already leaving like the low areas of Miami Beach. You see old railroad track actually, the old Flagler railroad from way back when, because that's the high ground.

If you have an area that has got a lot of risk, it can really start to go spiral down. And we call them climate abandonment zones where people are leaving and prices are going down and it's not coming back. So the individual homeowner and counties and cities that collect tax dollars from real estate valuations have a problem to face, right?

John Mulliken:

I understand that you actually began to work with some of the real estate brokers and real estate agencies. Do you think that there's a version of presenting climate risk that will work?

Ed Kearns:

I think so. And I think right now, and this has been born out by the Zillow joined Realtor and Redfin and homes.com as one of our partners to release this information. And it was a highly publicized case before the holidays at the end of 2025 where Zillow took down the scores. They still had the pass through to the First Street sites that people could still discover, but it wasn't so easily discoverable. We had to do an extra click, and this made the national news and such. And that was born about because of this tension between the sellers and the buyers, right?

It changes from provider to provider, but getting this information out there one way or the other is really the challenge. It's get it out there and , in my opinion, let the markets do what markets do and make those adjustments. Because government can't be in the situation of picking winners and losers here and stuff like that, but there has to be a level playing field that the consumers have the same information that the real estate folks have and that the insurers have.

Mike Toffel:

Yeah. It does bring up this question of which sector is going to be most effective in addressing this. There's lots of reason to think that this could be a private sector issue. Buyers and sellers are private actors usually. They'd have a demand for this information. So you might imagine that that would just become part of a transaction due diligence process and we don't need the government necessarily or nonprofits, but then there's all these market failures and they're differentially powerful and disinformation campaigns and all that. So then maybe there's a role for nonprofits to come in and be release maps and another source of information.

But then of course, maybe there's a government role here in saying, "In order for us to regulate transactions to make sure that externalities are properly accounted for, we need to mandate the disclosure of this information." And Connecticut is taking an interesting role here, right? Connecticut's working with your data and trying to make that available to its residents.

Ed Kearns:

Yes, exactly. They have licensed our data in order to pass it on to the citizens of Connecticut so they can start to understand that what that risk profile looks like in their neighborhood, at their home, at their business. So, and they're coordinating this across multiple government agencies, also letting the business community understand this as well too. So I applaud them for taking this step, right? And just licensing the data is really the easy part of what their challenge is. Now their challenge is communicating this risk across the state so people could understand, like we were just talking about, some people are going to be surprised by this at first, right? And it's going to be a journey for, I think for a lot of the state. But I think by moving early, the sooner the state and its residents can prepare for this and its businesses can prepare for this and mitigate this risk, there's a lot of things that can be done to reduce the impact of extreme events that are sure to occur.

John Mulliken:

Yeah, that's great. Well, so this really leads into some questions about capital markets and policy. You're thinking about how do governments interact with different markets and with purchasers. I mean, you have some big institutions now like KKR and a number of other very large investors who are working with your data. They're acting in different ways than consumers or individual businesses. I'd love it if you could take us inside your view of a deal or a due diligence or a process. So we talked about managing assets, but I'm really interested in how investors are using your data and your models to construct a view of how they're going to go about sourcing deals that they might do, how they're using it to figure out whether buy or don't buy and how they're thinking about managing those assets.

Ed Kearns:

Yeah. What I've seen, again, it's fascinating the diversity of views from different companies that have... Sometimes they have similar management practices, but very different views on how they're going to address the risk across a portfolio. And they all have different tolerances for that risk depending upon if they're acting on behalf of other clients or they are maybe they're, like I said, maybe like a pension fund, they have some responsibility either to a government or to a company, to some fiduciary responsibility to manage these things.

And the approaches that they take as they're assessing our data are, there's some real similarities there, right? So they usually have a... Before they really dive in deep, they'll have a subset of their portfolio that they are familiar with, that they've already done some kind of due diligence on. Maybe they have some concerns about it for one reason or the other, and it's basically like, "Okay, well, show us what your tool tells us about these properties." And we're happy to do that. And so we'll do these measurements and these assessments like I just described earlier, and we'll sit down with them and we'll go through it with them and explain why things are being assessed the way that they are. And yeah, you usually can see the light bulb go off.

And one of the ways too, and one of our tools is you can just type in any address and you can see any location pop up. And what most people do, whether they're the CEO of a company or they're an analyst or whatever, is they'll type in their home address. And I've seen this a thousand times, right? So they'll type in their home address. And most of us have a pretty good idea around our home address and our neighborhood, particularly for flood, what streets flood, what corners flood during a heavy rainstorm. And that's the first thing that they'll do and they'll look at the five year or the 20 year return period and they'll go, "Ah." And you can see the light bulb go up like, "Oh yeah, I recognize. Yes, you are predicting where that would actually happen. Oh, okay." And that opens the door to going in deeper and deeper onto the other perils also too.

But yeah, so it's this combination of institutional assessment of a group and then also that individual human connection to a property or a house. And then when they can look at the tool and they can understand that, oh yeah, this is resolving the kind of risk I would expect in that place I'm very familiar with, then it opens the door to other investigations.

John Mulliken:

It's really interesting. You talked earlier about Fannie Mae and Freddie Mac and they're beginning to use your data. This whole question of financial disclosure requirements as the ebbs and flows, as different administrations change and so forth, but I'm really interested in how this affects you as a firm trying to get your information out there. Does the fact that the moment changes how much emphasis there is on financial disclosure and risk disclosure, does that make it easier or harder for you as you're trying to get this information out?

Ed Kearns:

In some ways, in a very odd way, it makes it easier and I'll explain how. So when there was a lot of the climate regulatory policy being implemented or discussed being implemented a couple years ago, we did have a number of customers coming to us because they were worried about being in compliance with the regulation, with those new disclosure laws, right? So it was very much a compliance mindset. I'm going to have to answer the mail, I'm going to have to answer to the SEC or whoever. I'm going to have to do my homework and understand the risk of my portfolio so I can explain it to the regulators. So that was like one personality type. The other personality type is maybe driven by the folks we were just discussing, like at say a pension fund with some responsibility to manage this risk over the long term so they can hit their targets for returns.

They have a different motivation, right? They're trying to make the most of what they've got and it doesn't matter what the regulatory landscape necessarily is, but they know what they have to do in order to preserve their investment. And so when the change of administration happened and a lot of executive orders that were around climate risk targeted at US Treasury or whatever, those went away. So a lot of that stuff got dropped. The customers that were coming to us from a compliance viewpoint basically faded away. And it happened a lot at a lot of banks throughout the US, you saw staff actually being reassigned within the banks from the climate risk groups. Okay, well, we're not going to do climate risk anymore. You're going to go work someplace else in the bank or whatever.

So we saw a big deceleration on the compliance front, but it opened up our bandwidth. We're about close to about a hundred people now at First Street and can only engage with so many people at once, but what it did, it opened our bandwidth to be talking with these other groups that had a different motivation and like we can go faster with them now. So and like I said, in a strange way it's made it easier now that the regulation has been like in the US has been set aside that that momentum has gone. The momentum is picking up now on the other side. So that'll probably seesaw back and forth, I imagine, over the next decades ahead, but there's more than one reason to be motivated about doing something about climate risk, that's for sure.

Mike Toffel:

Ed, are there parts of federal policy that are helping with your business and with your ability to get climate risk out there at this point?

Ed Kearns:

Yeah. Yeah. No, I think there are some very positive signs on federal policy, but some of them are coming out of FEMA, which is not strictly financial, right? But they have just released-

Mike Toffel:

FEMA being the Federal Emergency Management Association.

Ed Kearns:

Yes, but they also do hazard mitigation and they also run the National Flood Insurance Program. And so we had talked earlier about some of the issues with the FEMA flood maps and how the average everyday American understands they have flood risk, whether they're inside a FEMA flood zone or not, right? And they don't have the full answer. But FEMA, to their credit now, has just released their first flood map with Harris County where Houston, Texas is, that was hit by Harvey years ago, Hurricane Harvey, horrible flooding there. They have now put out draft maps that include the heavy rainfall. So this is a tremendous step forward for Americans being able to understand and be prepared for their flood risk in this country. So yeah, kudos to FEMA for that. It's going to take some years to implement this across the country, of course. It will take probably a decade or more, by my opinion, probably to have this change percolate around the country and all the different assessments and modeling that needs to be done.

But we'll get there. On the financial side, I don't think there's been any federal policies that have come out that have helped us, quite frankly, but there hasn't been any that have hurt us either. So in the previous administration, there was an executive order to assess the climate risk on the health of the banking system and the financial system

Now, with the change of administrations, that executive order was rescinded,

Mike Toffel:

So there was some movement pending at the SEC, the Securities Exchange Commission, requiring companies, I think beyond the banking sector to reveal climate risks, and that's been put on pause or maybe canceled altogether. Would that have helped drive demand for the information that you and your competitors are providing?

Ed Kearns:

Yeah, it probably would have. Again, from the compliance standpoint, they would have brought more people from a compliance viewpoint forward. Whether that would actually solve the problem that we have as a nation addressing climate change, it might help some, but yeah, like I said, it's created room for other players in this space that have different motivations than that. The other thing is that the stockholders in these companies have also woken up to this reality also too. And so they may not be labeling it as climate risk, but it is finding a home within the risk management structures within companies. The idea of materiality, the fact that this can be quantified now because of the tools of First Street and other vendors in the climate services space have produced really encourages, I think, investors to demand. They want to know, how is this impacting things?

And I think as time goes on, even things like municipal bonds, which were always considered, "Oh, this is a safe haven, this is a sure thing." It's like, "Well, no, a lot of these projects are in some risky areas and what is the climate risk for that project that's being paid for by that muni bond? Is that a good investment as we think it is?" So these questions are being asked now and it may not just be like a requirement from the government, but it's just good business, right? So investors are going to be careful with their money and that this information, now that it's knowable, the climate risk is knowable from more than one source, I think the ball is really rolling now.

John Mulliken:

When we first started talking, you mentioned that when you joined First Street, it was a nonprofit and now it really sits at the intersection of being a commercial business and being a public good. I mean, you're even a public benefit corporation. Can you describe what that is? What is a public benefit corporation? How do you think about this role sitting at the intersection of these two very important parts of climate change?

Ed Kearns:

Yeah. It was very interesting to see this evolution of First Street from a nonprofit trying to figure out how do we communicate about climate change and say, "Oh, I think we understand where we can actually do a lot of good for industry and for this country by going down this road." And it was sometime probably in the middle of 2023 that it dawned on us that this was the change that was happening. And because we were licensing this data, data we were creating to different companies and different government agencies in different states, really seeing the uptick of the use and the revenue that was coming in was we were going to be breaking even soon. So our donors were very excited about this, right? They're like, "This is fantastic." One of the reasons I came to First Street too, as a data nerd, I was also very interested to see if we could create and maintain data based upon its value to industry and not take any government grants, not take any of this public funds that usually research is based on, but can we pay for itself? And the answer is yes.

And then the donors said, "Well, and then now you've also proven there's a market for this." And nonprofits exist to fill a need where for profits aren't going to do it or governments don't want to do it, either way, but there's a niche there that nonprofits fit nicely into. But we had shown that this niche we were in is actually better fit for a for profit. So yeah, we spun out as a for profit. We kept our public vision with the public benefit company that yes, we are doing a public good. We're creating data that are useful for the greater good of our society and our businesses. If this is going to help propel good decisions that mean a safer society for US businesses and US citizens then fantastic, right?

So it's a very fun job to have, but now also too, then as that market is maturing to see as our journey now, we're about a little over two years into the for profit journey here, and we're seeing the markets continue to increase in size and increase in skill. And they want more different types of data and more sophisticated data. And it's great to see this evolution from like, how do we tell the story? To like, oh, the story is landing and they want more, right? They want more detail. They would need a finer grained response on those financial impacts. And it's like, yeah, and we can do that. So it's been a very fun journey.

Mike Toffel:

Great. Well, look, Ed, we've covered a ton of ground. One of the final questions I ask of our guests, the final question I ask of our guests is to give advice on resources that listeners who are interested in learning more can consult in order to actually dig in, whether that be websites or podcasts or conferences or newsletters or books or reports, what would you suggest are some top resources that such folks could go consult?

Ed Kearns:

Yeah, there's some wonderful government resources both at the National Oceanic Atmospheric Administration, NOAA, as well as NASA, as well as the US Geological Survey, EPA. There are some wonderful data sources there that really describe the basics of climate in a very useful way. Then when you've got companies like First Street, we put out a lot of reports because we take our data and we apply them to certain questions, whether it's real estate or insurance or whatever. When we put out, and we usually have webinars that go along with them too, but that describe our data applied to that particular subject of interest. And so you see quite a few of those types of things out there too.

But the insurance industry and reinsurance industry put out a number of annual reports that look at, usually with a global kind of coverage, global kind of lens, showing how extreme events are impacting their bottom line, their customers. And it's very eye-opening when you start to look across, away from the climate service vendors like First Street, and you go to more of a general risk group, big reinsurance company, and you look to see how those risks from extreme weather events now are as big or bigger than anything else that they're worried about. This is not a niche subject anymore. Climate risk is not a niche subject. It is the subject across risk management now. Physical risk management.

And so it's very interesting. I encourage your listeners to seek out some of these, whether it's like a big multinational, like a Munich Re or somebody like that, go check out some of those reports. They're public and they describe the situation well. And watch out for those graphs with a lot of graphs showing increasing losses over time, whether it's from severe convective storms and hail, which caught everybody surprised by last year.

Mike Toffel:

It's all good. Great. Well, thank you so much Ed for spending time with us.

Ed Kearns:

Yeah, thank you. It's a pleasure being here.