Stormy Weather: Disaster Risk and the New Shape of Credit Transcript

KRISTEN BROADY: Good morning. I'm Kristen Broady, Director of the Economic Mobility Project here at the Federal Reserve Bank of Chicago. Thank you for joining us for "Stormy Weather-- Disaster Risk and the New Shape of Credit." We know that many factors impact access to credit and loans for individuals and businesses. Today, we're going to look at two of those factors, potential borrowers' credit profiles and geographic risk of flooding, wildfires, and other weather-related disasters.

Questions we'll consider include, what information do banks need to determine potential financial losses from disasters? And how do they adjust their portfolios when these types of disasters occur? You'll hear from Chicago Fed Vice President and Director of Financial Research Ralf Meisenzahl, whose research investigates whether financial intermediaries adjust their portfolios in response to multiple types of weather-related disaster risk.

Following Ralf's presentation, Bloomberg's Stacey Vanek Smith will moderate a panel of experts to address topics including the integration of weather-related disaster risk into lending assessments, policies that could mitigate potential credit deserts in communities that are vulnerable to weather-related disasters, and how collaboration between various groups can be fostered to address gaps in risk assessments.

And now I will turn things over to Ralf. I'm looking forward to his presentation.

RALF MEISENZAHL: Oh. Thank you so much, Kristen, for having me and having the opportunity to present to you. And thank you for the panelists for taking the time and engaging in the discussion. My name is Ralf Meisenzahl. And the main point that I want to discuss today is, how have banks, particularly large banks, responded to weather event risk?

So just to give you a little bit of an idea that weather events are in fact-- and disaster risk-- they are on banks minds, I will start off with this quote from regulatory filings of region banks that says, "The severity and impact of future earthquakes, hurricanes, severe tornadoes, droughts, floods, and other weather-related events are difficult to predict." And then they go on talking about what that means for their risk management.

Now, we also know that over the last decade, at least, extreme weather events have become more frequent. And they have also become more costly. And we want to ask, in the research that we conducted, what are the consequences for the bank's mortgage market? What are banks doing?

Now, I want to clarify before I show you any results that these results are conducted only with the portfolio of the about 30 largest banks. So things I will have little to say about is how smaller banks and community banks, or credit unions, or non-bank lenders have responded. But since these banks are the largest banks, we are concerned about their overall impact, because they are large lenders.

Now, with that, I want to start off with showing you that there is significant variation in flood risk. And here, I have a map, mostly along the Mississippi, just to highlight that flood risk is not just a coastal phenomenon. You also have it from river overflow from rain. And the more red you see in this map, the more flood risk there is. Green would be less flood risk.

Now, you can also already see-- perhaps you see the county you're living in. And you may wonder, how can I have that much or that little flood risk? This is a measure aggregated to the county level. So I'm going to work with fairly aggregate data. Here, you see it for flood risk. I've also done it for wildfire risk.

And one reason is that the time period I'm considering-- so I'm going to go back to 2012 and 2014. We did not have very granular data, necessarily, on the individual measures, that any particular house is affected. And when you talk about things like wildfire, it's also not clear what precisely banks take into account.

So some insurers, for instance, said that wildfire risk, they consider five-mile radius. Even so, some providers only consider a half-mile radius. So this is going to be an aggregate measure that I'm using. And this aggregate measure, as you can see in the map, is varying a lot across states. So you have a lot of green in Kentucky, a lot more red in Illinois. And you can also see that in a sea of red in Northern Illinois, there is a green dot. So we can use these differences, this variation in risk, to identify the effects.

Now, the measure I'm using here that comes from FEMA does take mitigation efforts into account. So it is not just a risk, per se, but also, is there a dam? How well maintained is this? Is the community resilient? So it's a more complex measure. But I just want to point out that this is taken into account. And again, don't think about this, what does this mean for your property, necessarily, in the county, but what does this mean for the county as a whole? Because we think that is mostly the unit of observation lenders consider.

Now, the first thing I want to show is, overall, an analysis of mortgage portfolios of banks. And we ask, over time, how did lending evolve? And how did holdings of mortgages of banks evolve over time, depending on the exposure to flood risk?

You can see, in 2014, there's a dot. That's a normalization. It's set to 0. So everything that I show you here is relative to 2014. And I picked 2014 mostly because it was the time of the Paris Accord. So that seems to be a natural normalization when you talk about event risk.

I also want to point out that you can already see this decline starting 2015, '16, '17. And you get fairly large changes. So a one-standard-deviation increase in the flood risk reduced the holdings of these mortgages by about 10%, which is a very substantial part if you consider this is a seven-year time period. So it's a 10% reduction. And it predates the large increases in property insurance premia that you may have recently read about in the newspapers.

I, in my research, wanted to be thinking about who is going to be most affected by this reduction in credit supply to areas with more flood risk, the same picture you can produce when you use wildfire risks. So this is robust. It's a robust finding across different measures of extreme weather risk.

And so I'm going to split the sample by borrower risk, which I'm going to proxy by the credit score. I will split the sample. And then you get a picture like this. So what you can see in this picture, and what it conveys, is that the groups with the lowest risks, meaning the highest credit score quartile, have the least reduction.

So this is relative to the second lowest credit score group. And you can see that the low-risk borrowers, they have actually still gotten more credit than the low-risk credit score borrowers. So there is a redistribution going on, or a differential effect going on. It is, in particular, the lower-credit-score borrowers that are affected by this change.

Now, if you think about who are these low-credit-score borrowers, that often includes lower-income households. It often includes first-time homebuyers, because usually, your credit score really gets high once you've taken out your first mortgage. So first-time homebuyers, low-income households are most affected by this retrenchment in credit.

Now, this, of course, will lead, I hope, into the discussion that we will have later in terms of what can be done. You can think about a second margin. Here, this is the extensive margin. Do people get loans? Do they have access to mortgage? But the second margin that you can think about is, how much credit do they get?

So we think about the loan-to-value ratio next. So the loan-to-value ratio tells you how much down payment you need to have. The lower the loan-to-value ratio, the more down payment you need to have. So we run the same regression. And we will look at new mortgages. And in this picture, you can see that in particular, in 2017, '18, and '19, higher down payments were required, meaning lower loan-to-value ratios, in areas with higher flood risks.

So banks not just pulled back in credit provision on the extensive margin, meaning the number of loans, but they also pulled back on the so-called intensive margin, meaning how much credit you get or how much money you have to put down to buy a house.

Now, you see that this has a U-shape, a little bit. And I just want to remind you that 2020, and of course, 2021, is an era-- a time of very low interest rates. And it has COVID. So it's not entirely conclusive whether this trend is going to continue now that the interest rates have gone up again.

Finally, I just want to also understand whether all banks are doing this, or which banks are doing this. And it turns out that not all of these large banks have reduced lending equally. In fact, the banks that reduced lending most are those that had little capital. So why is that? Capital is one measure of loss absorption capacity. So if a bank has little capital, it is less able to absorb large losses.

Similarly, I find that banks that had a high concentration of mortgages in their loan portfolios tended to retrench more. Again, the argument here is this makes sense because if you have a lot of exposure to the mortgage market, you want to reduce your exposure. And you most likely want to reduce your exposure in the riskiest part of your portfolio, which would be the low-income borrowers, the low-credit-score borrowers in the risky area, because they have the least ability to withstand these types of shocks.

And last, I also show that banks that generally, in their mortgage portfolio, have more exposure to flood risk, but also to wildfire risk and other measures, they also reduce credit. So you can think about this as generally a risk of situation, where banks that are more exposed to these risks have reduced their exposure to the risks more, and also have done this if they have less loss-absorbing capacity.

So in conclusion, what I've shown you is that large banks have retreated from high weather event risk areas. On the example of flood risk-- again, it holds with wildfire risk as well-- the brunt of this is borne by high-risk, low-credit-score borrowers that experienced the largest reduction in credit supply from large banks.

And again, I want to reiterate that these results are only for large banks. It could well be that community banks have stepped in, or non-bank lenders. But at least the largest banks have reduced their exposure to these types of loans.

And if you take this together, you find that this means first-time homebuyers, low-income homebuyers, will face additional challenges going forward in buying homes, and not just because getting the loan, but even if they get the loan, they probably need higher down payments, which is hard to do.

So I want to stop here. Thank you for your attention. And I'll hand it back over to Kristen. Thank you very much.

KRISTEN BROADY: Thank you so much, Ralf. I now have the pleasure of introducing our esteemed panelists, who are going to have a good time, I'm sure, discussing what Ralf just presented. First, Nicole Elam is the President and CEO of the National Bankers Association. Mac McComas is the Senior Program Manager of the 21st Century Cities Initiative at Johns Hopkins University.

Jenny Schuetz is Vice President of Infrastructure and Housing at Arnold Ventures. Nitzan Tzur-Ilan is Senior Economist at the Federal Reserve Bank of Dallas. And I now have the pleasure of introducing Stacey Vanek Smith, Senior Story Editor for Bloomberg Audio. Stacey, I'll turn things over to you.

STACEY VANEK SMITH: That was a really fascinating presentation and, I think, very timely, for sure. I wanted to start, I guess, with maybe a more basic question-- quite basic, actually. I did get Kristen's permission to ask such a basic question. But when I first saw the research and heard about what the topic was, I was really surprised, because to me, initially, it did not seem like the weather should really impact people's ability to get loans or businesses' access to financial services. But it clearly really does.

And so my first question to our panelists would be, maybe, why is that? Why is that relationship there? And also, what is at stake here? Why is this a relationship that is a good thing to look into? So I'd love for you all to jump in and just answer my very basic question.

NICOLE ELAM: Well, I love your basic question, because it seems counterintuitive. The reality of it is that weather should not impact your access to financial services, but it absolutely does. It impacts because, as a borrower, you're considered financially risky. Banks can't underwrite mortgages or small business loans in areas where roads are flooding regularly or there's no reliable water infrastructure.

Why? Because you may not be able to repay your loan if the weather destroys your home or your business's ability to open or make money. And so these loans represent not just financial risk, but potential generational wealth loss. And that's really what's at stake here. If it's harder for you to buy a home, if it's harder for you to build a business, then it's harder for you to create wealth.

STACEY VANEK SMITH: Thank you. Oh, yes. Please, please, follow up.

JENNY SCHUETZ: I just wanted to pull out a little bit that the financial channels through which climate and weather events impact banks and households and businesses differ a little bit, depending on the kind of weather event we're talking about. So it's worth pulling some of those out.

Ralf's presentation focused on floods. One of the things we know about floods is that they are really damaging and expensive to buildings and structures. And so they're one of the biggest sources of financial losses through climate. Wildfires also burn things down. They're very bad for properties. They cost a lot of money to rebuild afterwards.

So those are two of the climate risks we focus on most often. But we should also think about things like heat, which tends not to affect buildings that much, but has a big effect on people, so on people's health, has impacts for public health, on people's labor productivity, especially for people who have to be outside or in un-air-conditioned buildings. That impacts their labor, their ability to work, the number of shifts, their productivity.

And all of these things also translate into things like household credit. Whether you can pay off your credit card depends on whether your hours got cut because there was a heatwave. Small businesses see their income fluctuate. Restaurants that rely on outdoor seating have fewer people when there are bad weather events. Local governments spend more money to repair after disaster. And it may impact their intake from property taxes.

So there are a lot of actually quite complicated channels by which climate impacts people, buildings, businesses. And that goes upstream, then, to the kind of lending decisions that banks are making.

STACEY VANEK SMITH: Oh, that is really fascinating to me. While Ralf was talking, it actually struck me, one of the things he said about banks, after a disaster, just having less capital and less of an ability to take risks. And it occurred to me that that could also put-- to your point, Nicole, about generational wealth and buying property, it seems like if the banks have less money, they're probably going to not be so willing to take a risk on someone or give them a home loan if they're not-- don't already have all the money to pay it back. And so it feels like that can have ripple effects going forward for a long time.

MAC MCCOMAS: Yeah. And jumping in on that point, it also has an impact on local public governments. So in a 2020 paper by some of my colleagues, they studied the impact of Atlantic hurricanes on local public finance dynamics since the 1980s. And they found that the shock of hurricanes reduced tax revenues and expenditures and increased the cost of debt for up to a decade following exposure to a hurricane. And they found that, obviously, major hurricanes had a much larger impact than minor hurricanes. So there was a really long-term negative consequence to the borrowing capacity of public governments.

STACEY VANEK SMITH: Oh, that's fascinating. Thank you all for jumping in on that question. Well, I guess, moving forward in a certain-- maybe slightly solution-oriented ways-- what is a good way, or how should banks integrate the risks of big weather-related events and disasters into their lending decisions? Like, what would a better way be than what we're seeing now?

NICOLE ELAM: I would say that climate risk assessment is vital. But for minority depository institutions, these are banks that are predominantly owned or operated by people of color and predominantly sit in and serve communities of color. Integrating climate risk is not a choice. It's a necessity because of who they serve. Minority banks are located in and lend to communities with above-average climate risk.

60% of the zip codes in their lending portfolios have above-average climate risk. 46% of their branches are in high-heat-risk zip codes. That's more than twice the share of non-minority bank, community banks. And so minority banks are indeed on frontline lenders in these climate risk zones. But how do they do it? To your question, how do they integrate it?

They do it by accounting for both the environmental risk and the social impact of pulling capital from high-risk areas. They have to account for infrastructure realities in that community. They have to account for things like insurance access. How accessible, how affordable, is insurance? They have to follow insurance trends and guidelines, while also advocating for insurance reform, so that the products remain accessible and affordable.

But I think what the real headline here is that too often, integrating climate risk leads to withdrawal from the communities that need lending and access to capital the most. And so integrating risk can't mean retreat. Instead, it has to lead to innovative lending. It has to lead to infrastructure improvements. It has to lead to policy that reflects both the physical and the social risk. You have to have public-private partnerships that mitigate risk rather than avoid it. So integrating climate risk isn't just about the climate risk. It has to take into totality the entirety of the situation in the community that you're serving.

JENNY SCHUETZ: And ideally, what we'd like to see, not just with financial institutions, but with the actions of people, and of businesses, and of local governments, is thinking about the risks that might occur and then taking proactive steps to make yourself more resilient before the disaster strikes. We know that in communities like Texas, for instance, we saw reluctance to put into place sirens and early warning systems and to move cabins to higher ground.

So we didn't take the investments upfront to protect people. And then you wind up spending more money on the back end, doing the search and rescue, doing the repair, doing the rebuilding. So we want to figure out ways to align financial incentives so that people are making decisions to invest upfront, that banks are providing capital upfront for things like moving your-- elevating your house or moving things to higher ground, putting in insulation, so that people will be safe on the front end.

STACEY VANEK SMITH: I did not realize about the early warning systems, that that had-- the decision had been made not to put that into place. And that seems like it feeds into what Nicole is saying, that it leads to bigger losses and then retreat, as opposed to investing in early warning systems. That is really fascinating.

What about-- I'm wondering, what are some maybe metrics that could be used, that financial institutions could use, potentially, to compare the risks for weather-related events and disasters, maybe across different regions and communities?

NITZAN TZUR-ILAN: So let me jump here and say, first of all, the disclaimer, those are only my views, not necessarily the Dallas Fed views. But I believe that access to data is very important in order to compute the real risk exposure. And in order to do it, we need to think about climate risk across different regions, across different financial institutions, different climate events.

And we should have one matrix that is standardized, and transparent, and forward looking for future climate events. So we are currently using two different indexes from First Street Foundation and CoreLogic. Those are different physical risk score metrics from 0 to 100. We have these metrics for different climate events. So you can think about fires, floods, hurricanes.

And then we have an aggregate measure for each housing unit in the US. And this is based on different parameters. So we have historical information on historical climate events and also the damage that's caused by the climate event. We have also future and historical adaptation and future climate projections. It's very sophisticated and detailed models for overall climate exposure in the US.

This is on top of access to FEMA flood maps or any local climate information, including different climate events that we didn't used to take them into account, such as hail, wind speed, pollution, extreme temperature, et cetera.

I think that overall, it's very important to have access to those detailed data sets and to have one or two, a few similar data sets across different institutions, and that borrowers could also have access to those data sets. But it is very costly. And it's hard--

[AUDIO OUT] 

STACEY VANEK SMITH: I think we may have had a little connectivity issues, as happens a lot in our modern world. Does anyone else want to jump in about the data or the current metrics that we are using?

MAC MCCOMAS: Yeah. So one of the major issues is that the climate issue-- the climate models that we have for cities, where the majority of people live, and the highest-value property is located, they're currently very bad at predicting neighborhood-level risk exposure.

So one of the largest research projects that I work on is the Baltimore Social Environmental Collaborative, which is part of a new research program called the Urban Integrated Field Labs, which is funded by the US Department of Energy.

And one of the major reasons that this project was created four years ago was to improve climate modeling at the urban scale. And before this program, almost all of the climate models that exist were created in rural and suburban environments. And when the assumptions that exist in those models are brought into urban environments, they break down, because cities are just much more complex physical environments, with larger amounts of impervious surfaces, complex systems, such as water and sewer drainage systems, that make things like flood modeling extremely difficult to do. So--

STACEY VANEK SMITH: Much denser. It seems like that could compound the losses.

MAC MCCOMAS: Yeah. So we've been working over the past four years in Baltimore and with other field labs in Phoenix, Chicago, and Port Arthur, Beaumont, Texas, to improve urban-scale modeling so that we can better understand this climate risk in cities. But at this moment, a lot of the climate and risk models that currently exist make it very difficult to understand and predict future climate risk metrics in cities.

JENNY SCHUETZ: One specific example of what Mac is talking about, these overlapping systems-- so if you think about, for instance, a heavy rainstorm, like New York had last night, you get a lot of intense rainfall. And you may be able to predict roughly where the rain is going to come down most heavily.

But to figure out where the flooding impacts, you also want to look at things like whether you're on a slope. So if you have a building on the top of a hill, it's less likely to get flooding in the building than the ones at the bottom of the hill. That's something we know and can build into these maps.

But then you also have things like, what's the condition of the storm sewer system underneath the streets? And if the city either built this for a much smaller rain flow, or if they haven't been keeping it maintained and clean, it's the interaction between the physical risk, the amount of rainfall you get in a period of time, and the quality of the urban infrastructure underneath it.

And very often, that's not something that even cities have mapped out. Can they tell you where they have smaller and larger pipes to deal with the stormwater runoff? Because the interaction between those two is really what determines the impact on neighborhoods.

STACEY VANEK SMITH: And Nitzan, welcome back. Oh, yes. Oh.

NICOLE ELAM: Sorry about that.

STACEY VANEK SMITH: You've joined us again. Nicole, please make your point. And then, Nitzan, if you wanted to finish what you were saying, please jump in and do that.

NICOLE ELAM: I think Nitzan started with a great point, talking about how access to data is-- before you can even talk about standardized metrics, you need to ensure that you have access to climate data tools, and not just the access, but as the last two panelists pointed out, Mac and Jenny, where are the gaps in that data? Because there are a lot of gaps in the data that is out there.

But I think, once we have access to the data, and you understand the gaps, you also really have to pay attention to whether it's going to lead to exclusion and not inclusion, because I think, too often, standardized metrics can lead to exclusion and not inclusion. And so you have to make sure these metrics not just reflect climate exposure, but they reflect equity.

If you're not accounting for historic disinvestment, if you're not accounting for community vulnerability, if you're not accounting for infrastructure disparities, the metrics could lead to a new form of redlining, where you're essentially saying, oh, this neighborhood right here, I'm not going to lend to.

And so that's why you've got to make sure you've got the right people at the table, helping to shaped the development of these standardized metrics, so that they're accessible, they're usable by community banks, and they account for community vulnerability, not just the physical hazards. If not, you're going to punish banks who are doing the hard work of serving the hardest-hit weather communities. And you're also going to redline those communities at the same time.

STACEY VANEK SMITH: And Nitzan, I don't if you wanted to finish your thought or had an extra point to make.

NITZAN TZUR-ILAN: Yeah, let me say one more sentence, that I agree with all the rest of the panelists. I think that it's important to have one standardized metrics to think about future climate risks and historical climate risks that is going to take into account different factors.

But it's also important to highlight that it's very costly. And it's also hard to understand, because if you think about even the climate risk index, from 0 to 100, what exactly 70 means? Like, should I live in a place that has the overall climate risk of 70, if it's higher than 50?

So I think that financial institutions should hire climate analysts to think about those models, but also the public. If a borrower has access to Zillow or any other platform that includes information about climate risk, and then he's going to negotiate on his mortgage terms, how should we think about this information, and how the banking system should think about this information? So it's very complicated, but I think we should start to think about it.

STACEY VANEK SMITH: Well, speaking of that data, what about the banks themselves, the data that they have on how weather events impact their lending and what their practices are in these situations? Maybe, what kinds of data should they be disclosing for, I guess, the sake of transparency, and then also for the sake of helping these assessments be more accurate and more fair?

JENNY SCHUETZ: Those are two really tricky questions. It seems pretty clear that banks should be disclosing the kind of data they're using to make their underwriting decisions. So if your ability to get a mortgage or a small business loan depends on the climate risk measured at the property level, or the neighborhood level, or the county level, that's something that people should in advance.

And so disclosing with some transparency what they're considering in deciding whether to make these decisions is really important, just like people want to know what goes into their credit score. And this is based on, am I paying my bills on time?

I don't think it's banks' responsibility to disclose the actual climate data that they are using. They are consumers of this. They're purchasing this, likely from private providers. This really gets to the question of, whose responsibility is it to share information about localized climate risks, given that it gets used by banks, by insurance companies, by developers who are choosing where to build? And consumers, individual households, and local governments are often the last ones to have access to this.

And so we've typically relied on the federal government to be the data provider for information we think everybody should have available without having to pay for it and to have the documentation and guidance on how to use this and what this means. But this is also a new space. The federal government is often relying on data that comes from private companies to actually understand what the risks look like.

And so I think we haven't quite figured out, is this a new census product that should be put out? Is this something that EPA puts out and maintains over time so that people can track this information? That's a really difficult, complex question that has some ethical implications, as Nicole was saying. You wouldn't want to put out public data on a federal government website that contributes to, essentially, credit being withheld from vulnerable communities. And I think, at this point, everybody's been kind of punting on who owns the ball. I wildly mixed my sports metaphors there.

NICOLE ELAM: [LAUGHS] I think Jenny makes a lot of great points. I sometimes don't think that disclosure of the data is important. And I say that not thinking that-- not moving against transparency. Transparency is important. But I sometimes don't think that access to the data is the biggest thing, because, as Nitzan called out, even if you give folks the data, they don't know how to interpret it. What does a 70 score mean? What does a-- so people oftentimes don't know what the data means.

What we have found in our banks, what is most important is the practical ways that you are supporting that community. How are you helping them prepare for, in advance, some of these climate risks? How are you helping them engage with in retrofits and take advantage of grants and different local programs that may be available? How are you supporting them?

I think about Liberty Bank. In 2005, when Hurricane Katrina happened, what we found is that the biggest thing that people needed was access to physical cash. They needed physical cash to buy clothes, to buy food, to get hotels. So Liberty Bank had to create partnerships with other banks so that their customers could go in there and get access to capital, or to cash. They had to increase ATM limits so that people could get greater amounts of cash.

Sometimes they were lending without even knowing-- or providing cash without even knowing if the cash was there, because systems were down. So I think a lot of it is the practical ways that you're supporting these communities before, during, and after a disaster, and not just about inundating people with data that they may not even understand.

STACEY VANEK SMITH: Is it possible, though, that banks and financial institutions disclosing that data would expose maybe some of the unfair practices that you were maybe talking about earlier, like communities that they're not lending to, or maybe people that they're not lending to? I mean, it does seem like that might be a result of the transparency, potentially.

NICOLE ELAM: Yeah, there is benefits to disclosing the data. But it's, what are you going to do with it? I think about places like Chicago. Chicago uses their deposit power to incentivize good behavior from banks. So if you are a bank, and you want municipal deposits, you have to disclose where you're lending. That, I think, encourages investment.

So what are you doing with this data once it's disclosed? Is it being used for exclusion, or is it being used for inclusion? And I think Chicago is a great example of using the data to support inclusion. If you're tracking the data, and I'm seeing that you're not lending in certain communities, well, guess what. You don't have access to the municipal deposit, or you don't have access to these public funds. So I think that that is something, in a good way, that you can use data to incentivize better behavior.

STACEY VANEK SMITH: Oh, I love that. Oh, and Jenny, were you wanting to-- did you have a point you wanted to make?

JENNY SCHUETZ: I was just going to throw out there that it's hard to talk about disclosing data and regulating financial institutions without mentioning the Community Reinvestment Act, which has traditionally been the biggest federal lever for overseeing where banks are collecting their deposits, where they're making loans for mortgages and small businesses.

I know that in the recent rewrite of the CRA, there was a lot of conversation about how climate risks should be taken into account, whether banks should be given credit for making, for instance, resilience and adaptability investments, whether that should be on the credit side of the CRA ledger, or whether they should be held to account for pulling back capital in communities that have higher climate risk.

There is not going to be an easy answer to this. There are a lot of complicated ethical issues about disclosing information that could impact property values and people's well-being, or withholding information that could allow people not to put themselves at risk. So this is not a question that's going to have easy answers on it. And I think that's part of the reason that we haven't reached community agreement on what the right policies look like.

STACEY VANEK SMITH: I'm curious about the stakes of this issue. If we're looking at banks' lending decisions, maybe after a disaster or after a big weather related event, how can those lending decisions impact those communities and areas?

And I'm thinking about both property values, like you were just saying, Jenny, but also maybe larger economic development and the economic futures of those communities? What do you all see as the stakes with this issue?

JENNY SCHUETZ: I mean, there-- we don't even have estimates for the amount of money it's going to take to make communities more resilient. But people who have lived through natural disasters know what a difference it makes if you have funds afterwards, as Nicole was saying, for people just to get cash to be able buy food, and get back into their house, and buy cleaning materials, and clean out the house, and start over again, or for small businesses to replenish the stock that they lost.

If they can't have short-term cash, they are not going to be able to rebuild to the point that they were before. On the other hand, we have these longer-term questions of, do we want banks to continue making mortgage loans to new buyers moving into areas that are at high risk? Should people be informed of that and be allowed to choose to move to safer places?

What do we do about people who are living in places where risk will rise over time? Do we have provisions in place for people to move to safer places, for the federal government to buy out properties and say, this is not a place that people should be living in the future? Local government decisions about where to rebuild roads and bridges after disasters and whether to keep rebuilding them in the same places over time-- these are all going to have enormous financial consequences.

And you have to make a decision one way or the other. So whether you rebuild in a safer way or choose not to rebuild and relocate, both of those are potentially viable options. But they have really big consequences for the people who are currently living in communities.

NICOLE ELAM: Building off of Jenny's point about, what does it look like in a community post-disaster? We could look at New Orleans and see what happened. In predominantly Black communities, where lending dried up after Katrina, property values have remained flat for the last two decades, since Katrina in 2005, while there are other parts of the city that saw fourfold increases.

And so what we know is that a reduction in lending depresses values. It discourages rebuilding. And when there's no rebuilding, jobs and people don't come back. We see it time and time again. When a bank pulls out, property values stagnate, or they fall. Insurance becomes harder and harder to obtain. Intergenerational wealth building sells. So it really becomes a cycle of disinvestment when lending is pulled out of those communities.

NITZAN TZUR-ILAN: So let me jump here and follow up on Nicole's comments that my concern here is it's basically going to affect inequality. And the ones that are going to be affected more are low-income households, more minority populations. It's very in line with a lot of research that you just presented. Those are the type of households that, even before the climate event, they had a harder time to get access to credit.

Another point that I wanted to make is about the insurance market. Post-disaster, we see an increase in insurance prices. If you don't get insurance, you don't get a mortgage. So the fact of the increase in insurance is definitely going to affect households. I have a paper that shows the effect on household delinquencies, and also, Dawn [INAUDIBLE] from the Chicago Fed also presented work about the insurance market and the effect on households. I think, if you need to have access to credit, but you also need to have access to insurance.

STACEY VANEK SMITH: I'm curious what the research showed about, I guess, insurance markets and access to credit.

NITZAN TZUR-ILAN: So insurance premiums are increasing in recent years, especially after 2021. And there are a bunch of papers that already showed, using detailed data, that there are deserts in terms of access to insurance. Some insurance companies are pulling out from those high, risky areas.

And it also affects households. So we see that not just the increase in interest rate, that specifically, the increase in insurance premium is affecting households. And they are becoming delinquent on their mortgage. Some households are prepaying their mortgage in order to avoid the increase in insurance premium. So I think-- this is for a different panel, but I think we cannot avoid talking about insurance market when we are talking about access to credit in those high-risk areas.

NICOLE ELAM: Rising insurance premiums are the biggest thing that is hitting borrowers, particularly in low-income families. Lower-income families are paying the most just to stay insured. And so it is a huge cycle. That's why it's not just about lending. You've got to hit on infrastructure. You've got to hit on insurance reform. All of these things work together.

JENNY SCHUETZ: And I should say, this shows up in the rental market as well, that the landlords of affordable housing, subsidized properties, are having a really hard time getting insurance coverage and paying for the insurance premium. And again, to Nicole's point, if you can't get insurance for the building, you can't get a federally backed mortgage, particularly if there are subsidies involved with this. And so we are at real risk of some of the very scarce supply of affordable rental housing not being able to continue operating because they can't get insurance on the properties.

STACEY VANEK SMITH: Well, I'm curious about some policies that might help to address this, I mean, both what you were talking about, Nitzan, with the credit deserts, but also, Nicole, what you were mentioning about insurance premia going up so much that people can no longer afford to make their mortgage payments. Maybe, what are some potential policies or practices that financial institutions, policymakers could put in place that might help address this? Because it seems very key.

NICOLE ELAM: I think we need policies that redistribute risk, not just concentrate it on the backs of mission-driven community lenders. We just heard at the start of this panel about the research around large banks retreating. We've seen this time and time again. Large banks retreated from churches. Large banks-- and so you're seeing this. And we've got to find ways to redistribute risk because it's becoming more and more concentrated on mission-driven community lenders.

So things like loan guarantees and loan loss reserves for high-risk zip codes-- that can encourage lending. CRA reform, as Jenny mentioned-- that can encourage large banks to partner with, purchase MDI-originated loans. Public and philanthropic investments in weather-resilient infrastructure and retrofits can lower the risk for everybody. I'll maybe talk later about toxic asset funds and federal facilities that can absorb non-performing disaster loans. So I think these are some of the things that we can do from a policy perspective that can help redistribute risk.

JENNY SCHUETZ: Yeah. And one of the tricks we're going to have in the policy responses is that the level of government that has levers and authority over these is pretty different. So the federal government has a big footprint in regulating mortgage markets, and banks, and credit standards. And so that lives in the federal level. Insurance markets are, by and large, regulated at the state level. And so states have different kinds of authority over, for instance, setting what the insurance premium is, requirements for whether insurers have to allow people to renew their policies if they already have one, how much they can raise the insurance premium after disasters happen.

And states have taken very different approaches to this, including in providing the state-run fair share programs that are the backups to the private sector. A lot of the decisions about where you're allowed to build, building code, where you're not building, where local governments are investing in the infrastructure resilience, those are undertaken, and financed, and implemented by local governments.

So we don't have a ton of coordination between federal, and state, and local entities to get them to align their incentives on that. And that is really one of the key challenges, is making sure that everybody is on the same page, and they're rowing in the same direction.

STACEY VANEK SMITH: About to jump in there for a sec?

MAC MCCOMAS: Yeah. I was just going to continue and agree with everybody that, yeah, I think the Community Reinvestment Act was a huge opportunity. And the 2023 final rule did have-- was planning to adopt regulations that would give banks CRA credit for investing in climate change resilience and adaptation projects. But the current federal administration recently announced that they're going to withdraw that. So that's a pretty big loss there.

But even when you have these incentives to invest in adaptation and resilience, which I think we obviously need much, much more of, it becomes a really complex decision science question. So, for example, if a city, or a rural area, or a suburb knows that they're experiencing increased heat risk, that's great, that they know that and that they want to invest in it.

But there are a variety of strategies you can do to try to mitigate that. So you can plant more trees. You can invest in cool roofs. You can build community cooling centers. But all of these different investment strategies have different implications for different populations, different people. They cost different amounts.

So what you have is this problem of many different stakeholders, many different communities, and people impacted differently by these strategies. And so you need better tools and better ways of coming up with decisions on what to invest in and what kind of goals and impacts that you ultimately want to see.

STACEY VANEK SMITH: Seen any cities or states try anything that's worked particularly well? I mean, you were just talking about that, Mac. I know Jenny was mentioning, different states are trying different approaches. I'm wondering if there's anyone that's kind of trying something really interesting or has tried something that's really worked.

JENNY SCHUETZ: I'll throw out, as a-- so I joke in my work on land use regulation, we always punch on California. But California kind of brings this on itself. And so, for instance, if we look at the way California has approached insurance market regulation, they have tried to keep premiums affordable to people and so have capped the amount that insurers can raise premiums over time.

But as a result of that, insurance companies have been paying out more in damages than they have been taking in. And so we see major insurance companies leaving the state altogether. And so there are some limits on what you can do. And it's like, this is an issue that all of the state insurance regulators are going to have to grapple with. At what point does the public sector take in and assume the risk that the private sector can't afford? And how do you pay for that? And how transparent are you about taking on the risk and shifting the risk?

And so I think we have seen the states that have had the first exposure to this. California, Florida, Louisiana have all really seen their insurance markets under intense pressure. And so they are the first test cases on how to do this. And even there, we see really big differences. For instance, Florida has just expanded its public program much more and recognized, we're just going to own a lot more of the risk directly, but also then trying to think down the line about what the incentives are for contractors, for individual households, if they're primarily buying through the state.

NICOLE ELAM: I think one thing that is that if you try to put all the risk on one stakeholder, it's not going to work. And so while California's intentions were great, it led to the private sector leaving. And so places where you have fixed-rate insurance models, where seasonal or disaster-driven premiums spike, when they're offset by philanthropy, or when they're offset by state support, that works better. If you try to shift it all on one party, it's not going to work. And so I think it really speaks to the fact that you need everybody at the table. It's not just one lever. It is multiple levers that you need to be pulling to ensure that this is working properly for everybody.

STACEY VANEK SMITH: I mean, that is such an interesting kind of catch-22 situation, where the state steps in to try to protect its people. And then the business is so exposed that it leaves, that-- it's the retreat that you were talking about earlier. And there has to be a way to protect the financial institutions. But there also has to be a way to protect the borrowers.

And I'm wondering maybe what some solutions are to protect them that wouldn't also put financial institutions in a really difficult situation, like we saw in California. Maybe, what are some solutions for the borrowers themselves, to help borrowers in lower-credit areas that are maybe at a higher disaster risk? Maybe regulations or practices that could help protect them and not prevent them from owning homes or building wealth, and some of the things we were talking about earlier.

JENNY SCHUETZ: There are a number of different ways that we can think about how to help individual households. And thinking about people's homeownership decisions is one. So I do think, at some level, we're going to have to provide people with information at the point in time where they're choosing whether to buy a house and where to buy a house, so that they know the risk that they are assuming and can decide whether that's appropriate for them.

The National Flood Insurance Program, which covers essentially all of the flood damage for properties across the country, does require that if you're buying in a flood-prone zone, that you have flood insurance that covers you. And FEMA has been requiring that to rebuild properties in high-flood-risk areas, the properties have to be elevated.

So we're doing physical resilience. We're doing disclosure of risk. We're doing financial backstops that appropriately cover all of that, so all of those things aligned in the same direction. We do this more for flood than other kinds of risks because we have a more established of program for dealing with flood.

I think it's a really interesting question on wildfires, whether we wind up with, say, a national wildfire backstop that takes some of this off of individual states, and whether insurance companies are able to require fire resilience things, so putting different materials on the exterior of the building, clearing brush, getting people just good guidelines about, what are the physical investments that protect you against various kinds of risks?

What are the financial benefits of doing those investments? And what's the cost? So if you hire a contractor to fireproof your house, do you know how much you're supposed to be spending on that and how much you're likely to save over the long run? The cost-benefit information is something we could do a better job of providing to people. We just haven't really done that yet.

STACEY VANEK SMITH: Is that potentially putting more obstacles in place, though, for maybe would-be homebuyers who are maybe lower income or middle income? It does seem like it's adding expenses. Like, that does seem like an issue that may just be inevitable. But it does seem like it's going to raise the bar even more, maybe, for first-time homeownership.

JENNY SCHUETZ: I think that really depends on where we're looking at. And in some cases, the high-risk areas are also places that have lower incomes and more vulnerable populations. But that's not always the case. There are a lot of coastal locations. Think of places in Florida where there are a lot of second homes and a lot of wealthy people who are essentially choosing to self-insure so that they can live right on the beach.

So I think we wind up in these really different positions where, if it's a risky place that's occupied by relatively affluent people who are making informed choices, probably, there shouldn't be a ton of public subsidy for people to buy second homes on the beachfront, where taxpayers are going to have to pay to rebuild it.

On the other hand, people who are in places like Louisiana, where the flood risk has risen over time, people bought homes there 30 years ago. And the risk was lower to them. And now they don't have places to go. And the value of their property doesn't allow them to sell and move on. Those are really different cases for whether the taxpayers should be providing direct subsidies.

MAC MCCOMAS: One of the local examples that I like to look at is Philadelphia's Green City, Clean Waters program. And so this was back in 2011. They started a 25-year program to-- instead of investing a lot of money in one big gray infrastructure project, which I think a lot of cities and rural areas do too much-- building a big wall or investing a lot of money in tunnels-- they created incentives for businesses, nonprofits, homeowners, to invest in green infrastructure that they wanted, so things like buying-- incentives and grants for buying rain barrels or installing rain gardens, but things that would reduce property-level risk, but would obviously have benefits for the city as a whole when you distribute that throughout the city and reduce overall risk.

And so I think that's an example of where you can be flexible in creating a generalized pool of funds, but then let people themselves determine what solutions they think would work best for protecting their personal property or business or community organization.

STACEY VANEK SMITH: Maybe talk a little bit too-- we don't have a ton of time left. But I would love to talk a little bit about some of the incentives, as well as regulations that could be used, to help, I guess, encourage banks to adopt maybe longer-term strategies that would be more beneficial in the longer term, especially with regards to weather-related events.

I mean, are there, I guess, yeah, incentives and regulations that you all see as maybe being especially potentially effective?

NICOLE ELAM: Yeah. I think-- we've talked a lot about them, but loan loss reserve funds, loan guarantees, toxic asset funds, or federal facilities that allows the banks to get off their balance sheets. These nonperforming disaster loans, incentivizing lending for green infrastructure, CRA reform, housing retrofits-- I think all of those things are great policies.

A recent example is actually the Greenhouse Gas Reduction Fund. What I thought it did a great job was that it embedded equity into the policy. And so EPA awarded $27 million in competitive grants to mobilize climate projects, to mobilize infrastructure, to mobilize technology.

But it really incentivized and centered what they call the Justice40 Disadvantaged Communities when it was allocating resources, understanding that these were the ones that would be the hardest hit. And these were the ones that would benefit the most from it. And so I think, again, all of those things that help to redistribute risk, but policies that really embed equity into it, are also going to be successful models.

MAC MCCOMAS: As has been mentioned, I think states have a role to play here. And so this is not a climate-related example, but there's no reason a state couldn't do this for a climate-related focus. But the state of Maryland recently passed the 2023 Access to Banking Act, which offered an assessment credit to state-chartered banks and credit unions that operate in open branches in low- and moderate-income communities. And they established a fund to invest in new financial tools that would increase access to capital in those communities.

So again, this wasn't specifically weather or climate focused. It was more focused around general access to capital. But I don't think there's any reason why a state couldn't do something similar for a climate-focused fund.

JENNY SCHUETZ: And I'm going to play the economist here and say, we should also be really careful that we're not setting up incentives for unintended consequences. New York City is requiring all of their buildings to be net zero. So they're requiring property owners to retrofit and put in energy-efficient appliances, and insulation, and so forth.

The potential risk is that for low-income households living in subsidized rental properties, if you don't provide a subsidy to make that change, then the landlord has to pay a lot of money up front. And the only way they recoup that is by raising the rent. And so if we're going to require the regulation there, particularly on the low-end rental stock, we need to match that with subsidies so it doesn't just wind up displacing people.

STACEY VANEK SMITH: --turned her camera on. This has been such a wonderful discussion I feel like the time really flew. I definitely want to thank everybody and hand it back over to you, Kristen.

KRISTEN BROADY: Stacey, I just want to say thank you for joining us, again. This event was a follow-up to an event with Chicago Fed Senior Economist Dan Hartley. And so it's been great to have you for this follow-up. I certainly learned a lot. I hope that our audience did as well.

I want to thank Nicole Elam, Jenny Schuetz, Mac McComas, and Nitzan Tzur-Ilan for joining us today. This was really an enriching discussion. I learned a lot. I know that our audience did too. I've gotten text messages and DMs during the event with positive feedback.

So I want to again thank Ralf Meisenzahl for his presentation. We will be sending a post-event survey. So please be on the lookout for that. Your feedback helps us to build more events like this. You can find additional information about Ralf's work, along with work from other Chicago Fed economists, on our website, chicagofed.org/mobility.

You'll also find a recording of today's event there in a couple of days. We'll email you to let you know that it's ready. Thank you, everybody, for joining us. And enjoy the rest of your day. Thank you.