The Availability and Impact of Public and Private Funding Following a Natural Disaster

KRISTEN BROADY: Good morning. I'm Kristen Broady, Director of the Economic Mobility Project here at the Chicago Fed. And I'd like to thank you for joining us for the impact of public and private funding availability following a natural disaster.

Natural disasters, like hurricanes, tornadoes, earthquakes, and wildfires create economic shocks and impact income, wealth and capital.

According to federal statistics, over the last few decades, the number of natural disasters where losses exceeded $1 billion has been increasing. The availability of public and private funding after a natural disaster has the potential to mitigate some of those costs. Homeowners insurance enables homeowners to cover the cost of repairs following a natural disaster. But it's becoming more expensive, and some insurance companies are limiting plan offerings in some states.

In the aftermath of a natural disaster, community banks have the potential to provide liquidity to homeowners who are uninsured or underinsured. But community banks are less likely to have the capacity to lend after a large scale natural disaster, particularly if they are unable to raise sufficient capital. So how do community banks provide financial support following a natural disaster?

During today's event, you'll hear from Chicago Fed economist Daniel Hartley, who will present results from three of his studies. In Credit When You Need It, Dan and his coauthors analyze the impact of emergency credit access on households finances after a federally declared natural disaster. And in Natural disasters, local bank market share and economic recovery, Dan and his co-author look at differences in post-disaster credit allocation and regional redevelopment based on the concentration of local banks.

In Weathering an unexpected financial shock-- the role of federal disaster assistance on household finance and business survival, he and his co-authors study the financial impact of FEMA individual assistance grants in the wake of a tornado. Dan's presentation will be followed by a discussion on topics, including the impact of emergency credit following a natural disaster, insurance costs and state level price caps, and how the scale and spatial damage of various types of natural disasters necessitates differences in financial mitigation policies. We are very pleased that you have joined us for this event this morning. And now, I'll turn it over to Dr. Daniel Hartley.

DANIEL HARTLEY: Thanks, Kristen. Hello, I'm Dan Hartley, and I'm going to talk about research that I've been involved in, looking at how the availability of public and private funding affects households after a natural disaster. Importantly, these are my views, and not necessarily those of the Chicago Fed or the Federal Reserve System.

Next slide, please. So how do households recover financially after a natural disaster? Insurance policies provide the first line of defense. Some losses are covered by homeowners, renters, or auto insurance policies. However, damage due to flooding and earthquake is typically excluded. Insurance for flooding and earthquake is typically provided separately, mainly by the Federal government.

After a large natural disaster, a presidential declaration of disaster may be issued. This typically triggers aid for uninsured losses. The two main types of aid available to households are FEMA individual assistance grants, which are targeted toward low income households, and SBA household disaster loans, which offer low interest rate-sorry-- loans to creditworthy borrowers for repairing or replacing autos, homes, and home contents. Next slide, please.

I'm going to briefly discuss the results of three research projects that I have co-authored recently that shed light on how the availability of public and private funds affects households following a natural disaster. I'll cover the first study in slightly more detail and give a brief synopsis of the other two. The first, titled Credit When You Need It, is co-authored with Ben Collier, Ben Keys, and Xian Ng.

The second, titled Weathering an unexpected financial shock-- the role of federal disaster assistance on household finance and business survival, is co-authored with Justin Gallagher and Shawn Rohlin. Finally, Natural disasters, local bank market share, and economic recovery is also co-authored with Justin Gallagher. Next slide, please.

In Credit When You Need It, my coauthors and I estimate the effect of being approved for an SBA household disaster loan on household debt measures from credit records. We use almost random variation in disaster loan approval to measure its effects. Specifically, as the debt to service income ratio drops below 40%, the probability of approval for an SBA household disaster loan jumps by about 20 percentage points. Thus, slight and possibly random differences in debt to income ratios induce large changes in loan approval rates.

We focus on a sample of natural disasters, which occurred from 2005 to 2013, for which SBA household loans were made available. In this sample, the maximum loan amount is about $200,000, and the average is $42,000.

The average interest rate is 2.72%. Average maturity is 20 years, and average monthly payment amount is $229. If the loan is above about $25,000, then it must be secured by any available collateral.

Loan approval is based primarily on the applicant's credit score and debt service to income ratio. Only about 50% of applicants are approved. Next slide, please. While disasters have negative effects on both households that are denied for an SBA disaster loan and those that are approved, our estimates show that the causal effect of SBA Disaster Loan approval is to reduce the share of households that have filed for bankruptcy since the disaster by about 2 percentage points by 2 and 1/2 years after the disaster, which is similar in magnitude to the increased probability of bankruptcy due to the disaster.

Next slide, please. We find evidence that SBA loan approval is helpful in staving off a debt spiral. By two years after the disaster, the largest reductions in the likelihood of delinquency are for the most severe type of delinquency. We find a 5 percentage point reduction in the likelihood of having an account in collections. These estimates are consistent with loan approval helping to ameliorate liquidity problems that households face in making payments in the year following a natural disaster, which may result in accounts that are derogatory or in collections by the time two years have passed since the disaster.

Next slide, please. Among households that have credit card, mortgage, and auto loans prior to the disaster, SBA loan approval has the largest reduction in delinquency for credit card accounts. This result is consistent with prior studies that show that households facing liquidity problems stop paying unsecured debt first, since doing so will not jeopardize possession of their vehicle or home. Next slide, please.

Finally, we estimate that approval for a disaster loan actually increases the likelihood of taking on a new auto loan by about four percentage points. Presumably, the initial spike is driven by replacement of vehicles damaged in the disaster, but the persistence of the effect is likely due to the negative effects of the disaster for households that are not approved for a disaster loan. Rather than crowding out private lending, the liquidity provided by disaster loans appears to give households additional capacity to finance a vehicle.

Next slide, please. To summarize, we find beneficial effects of SBA disaster loans and reducing account delinquency, particularly for unsecured credit card debt, preventing debt spirals, which lead to bankruptcy and spurring new auto borrowing. Finally, we find evidence that these effects are driven by the liquidity effects of the disaster loan rather than the wealth effects due to the low interest rate of the loan, providing a subsidy to households. This evidence comes from examining another piece of quasi random variation, this time in disaster loan interest rates, which occurs with applicants near the 700 FICO score threshold.

Next slide, please. In Weathering an unexpected financial shock-- the role of federal disaster assistance on household finance and business survival, we use the randomness of tornado paths within counties to estimate the effects of tornadoes. We focus on a sample of severe tornadoes from 2002 to 2013. We compare the effects of tornadoes when FEMA individual assistance grants are available to the effects of tornadoes when they are not available.

The maximum grant is for about $30,000, but the average grant is in the $3,000 to $6,000 range. We find that low credit score households particularly benefit from grant availability. We find a four percentage point reduction in having a 90 or more day delinquent account for these households when FEMA grants are available, relative to when they are not available. Next slide, please.

Finally, in. Natural disasters, local bank market share, and economic recovery, we use variation in the concentration of local banks, as measured by their deposits, to examine the role that local banking plays in lending and economic recovery after a natural disaster. We find that the 75th percentile county, ranked by county local banking concentration, on average has 13 percentage points less mortgage lending than the 25th percentile county after a natural disaster. In other words, less mortgage lending and counties with more locally concentrated banking. We also find possible differences in economic recovery, pointing to lower population and per capita wage growth in counties with more locally concentrated banking, but these estimates are not very precise.

Next slide, please. To conclude, we find large positive effects of SBA disaster loans, beneficial effects of FEMA individual assistance grants for low credit score households, and less mortgage lending following natural disasters, particularly in counties with a higher concentration of local lenders. One caveat of this research is that we are unable to say much about whether the availability of disaster aid programs after a disaster induces households to purchase less insurance coverage up front. Thank you very much. Back to you, Kristen.

KRISTEN BROADY: Thank you so much, Dan. I now have the pleasure of introducing our esteemed panelists. Dr. Benjamin Keys is the Rowan Family Foundation professor of real estate and finance at the University of Pennsylvania's Wharton School and a research associate of the National Bureau of Economic Research.

Dr. Ralf Meisenzahl is a Vice President and the Director of Financial Research in the Economic Research Department here at the Federal Reserve Bank of Chicago. Dr. Amine Ouazad is an associate professor at HEC Montréal Department of Economics. And I am very happy to introduce Stacey Vanek Smith, Senior Story Editor for Bloomberg Audio and Special Correspondent for Marketplace. Stacey, I'll turn it over to you.

STACEY SMITH: Thank you, Kristen. What a fascinating paper. I feel like these are issues that have been coming up-- I guess they always come up, but especially recently. And I'm really thrilled to be able to take part in this discussion. I have so many questions, so I'm just going to dive right in, if that's all right. My first question, I guess, is about this emergency credit that the government can offer. And I'm really curious how access to that government credit can kind of help people's finances after a natural after a natural disaster. Ben, I know you co-authored the paper. Can you talk a little bit about this-- maybe some of the different findings? I S we just heard a little bit about them, but can you maybe dive in a little deeper?

BENJAMIN KEYS: Thanks so much, Stacey. And thanks to the Chicago Fed for having me. I'm biased and spoiled in that I'm a coauthor on this paper with Dan. But as we're looking at these federal disaster loans, this is a group of people who have uninsured losses after a disaster.

And so they may have had insurance that just didn't cover the kinds of losses that were ultimately created. They might not have known that they needed a flood insurance policy. We look at recent disasters like those in North Carolina or Vermont, where there was extensive flooding, and very few people had flood insurance. And so that's the kind of group that these programs target.

And what we find is a very large and persistent effect on receiving one of these loans in terms of households' financial situation. And I think the biggest headline result is related to bankruptcy. So bankruptcy is this extreme outcome, where you reach a severe point of financial distress. And we find that receiving one of these loans cuts the rate of bankruptcy filing in the following 2 and 1/2 years by almost 2/3, by about 60% So it's a really large benefit to receiving this type of loan. And I can go into some more detail if it's useful, but I think I'll stop there.

STACEY SMITH: I mean, that's a huge difference as far as bankruptcies. I mean, I'm wondering if you were surprised by that because it seems-- I feel like that's way beyond a significant difference. That's pretty stunning.

BENJAMIN KEYS: It's a big effect. And I think it points to just the liquidity crisis that these households face in the aftermath of an uninsured disaster. So put yourself in their shoes. You've had your home flooded. Maybe the roof of your house has been torn off. That is an expense that most people do not have the cash on hand for.

They might not have the liquid savings available for it. They probably can't put that on their credit card either. And so what you're then looking for is a large lump sum cash infusion to provide that liquidity. And I think what these results suggest is that in the absence of that, households really muddle along. They struggle to make their other payments while also trying to manage the repairs needed for their home.

STACEY SMITH: What about household finances in other ways, like maybe in the longer term? I know a lot of the paper followed people for quite a bit of time, not just immediately. I'm wondering what some of those findings were?

BENJAMIN KEYS: Yeah. I mean, one of the neat things about our, our project-- and I really commend the SBA for allowing us to link the administrative application data to credit record data-- is that we can follow these applicants for many years into the future after they've applied for these loans. And we find that these effects are quite persistent. And so even three, four, five years later, you can still see benefits of receiving this large cash infusion right after the disaster when households needed it most.

And to one of the results that Dan pointed to, we see just having any debt in collections, which is kind of a clear indicator that you've fallen behind. You've probably missed three or four payments, at least on a credit account. Those effects seem to persist even three or four years later. So it does seem like there are really long term benefits. On the flip side, we see more spending in the form of auto loans and taking on new debts going forward as well. So it looks like the effects are both quite large and quite persistent.

STACEY SMITH: What about employment outcomes? What do we see there?

BENJAMIN KEYS: So we don't have clear insights into to these households employment situations. I think it's something that we'd love to be able to access that additional data. That would be one more data lift, though, to bring that in. So at the moment, we're very happy to be able to have some, I would say, unprecedented look into the household's balance sheet. But I think the income questions are really relevant. And I think there's been some other work that's looked at this as well in terms of incomes. I don't if Amine wants to jump in talking about the effects of disasters on employment opportunities and migration.

STACEY SMITH: Yes, please. Amine, jump in.

AMINE OUAZAD: Yeah, of course. I think one of the ways that the funds can be spent is either as consumption or investment. And I think I'm a big fan of the work because I think the work really suggests that households are using these funds for in ways that smooth their consumptions, their debt, and allow them to overcome the obstacles that they're facing. And I think what we really see here is that there could be alternatives, like payday lending, that could be impeding the ability of these households to overcome these obstacles.

And I think this emergency credit, what it tells us is that households are not being preyed upon. They're not facing situations where they have to pay very high fees to overcome these challenges. But in terms of employment outcomes, to go back to the point that we were discussing, I think there needs to be a little bit more research on these questions.

I've published a paper on what is the impact of losing your job on a variety of outcomes, such as crime, such as long-term employment outcomes. And what we see is that when someone loses his job, the typically very long run impacts that persist for a very long time. And I think what the paper tells us is that, by providing credit, we can avoid these long-term impacts on employability, that those households are not being displaced, or they're not experiencing these long-term outcomes.

And so one way to think about the paper-- and again, I'm a big fan of paper, is to think in terms of the benefit and the cost of providing this emergency credit. I mean, it looks like there's very large benefits, and that the costs are relatively moderate. And so in terms of, you know, what bang are we getting for the buck, I think this seems to be a policy that's very effective.

And there were lots of conversations in 2008 with the great financial crisis about payday lending and predatory lending. There were a number of states that passed the laws or brought regulations on payday lending. And I think what we're seeing here is that this emergency credit is likely a better alternative than these more expensive forms of credit. So thinking about payday lending, if someone pays $15 of fees on $100 of loan, that's a 400% APR. And so it doesn't look much when it's $15, but overall, that can really significantly impede the ability of a household to stay afloat.

And so I think this is what I really like about the paper, is that it really gives us an interesting a cost-benefit analysis. And implicitly, there's the idea of what could households do if they didn't have access to these types of lines of credit? So yeah, I think, you know, the other way we can think about the paper is in terms of what kind of equity or balance sheet households have.

I own a home, I own a car. And if I lose my home or lose that car, I lose a lot of equity. And losing my car means I want to have access to jobs. And what I really liked in Dan Hartley's chart is that you can see that the mortgage and the auto loan, they have almost the same impact. The difference is not statistically significant.

That tells us that keeping my car gives me access to jobs. It allows me to go to job interviews. And in fact, if we look at the point estimate, the auto loan looks even marginally-- but I think here, I'm inferring too much-- more important than the mortgage in terms of point estimates, although the difference is not statistically significant. So that really tells us that households care about keeping access to job opportunities, that they're using the funds to adapt to these natural disasters, and that they're probably doing better than if they were the target of predatory practices.

STACEY SMITH: Well, I'd love to jump into that a little bit more. You talked about auto loans. And that was, I think, one of the most interesting points for me. Can we talk a little bit about how having that access to emergency credit in the moment affects things like household borrowing in the years after a disaster-- things like auto loans or even just household loans for bigger purchases. I'd love to break out a little bit the impacts we saw there.

AMINE OUAZAD: Yeah. I don't want to-- maybe Ralf wants to step in. There's the fact that the paper shows that there's increased borrowing after a natural disaster, and I think that we should see this as a sign of financial health because being able to borrow means that we have the collateral that enables us to take some leverage and to further invest in the quality of our collateral.

And when I talk about collateral, I talk about home equity. I talk about equity we have in our car, maintain that car, maintain that house. And we know that the insurance payouts are sometimes not sufficient to cover the expenses that will enable the house to survive the next natural disaster as, I think, evidence for a range of natural disasters suggest that they're more frequent. And so we could talk about different types, hurricanes, storms, wildfires. By "storms" I mean convective storms.

And I think here there's a parallel between rescuing-- and I hope I'm not saying anything controversial here but Main Street and Wall Street in the sense that here we're preserving the good collateral of those households by allowing them to get credit to maintain that good collateral. And I'm borrowing terms from the literature on the rescue of financial institutions, of banks, but here we're talking about the welfare of households. We're talking about their ability to stay in their home and to keep their cars.

So I think one way to think about the paper is to think in terms of helping them maintain this good balance sheet and this good equity. And like in the literature on financial institutions, there seems to be long-run impacts for a temporary policy.

STACEY: Yeah. Ralf, I'd love to-- if you have a point, something to jump in with, too, on this because there's a lot to unpack, for sure.

RALF MEISENZAHL: Well, there's a lot. And I just want to actually expand a little bit on the point that was just made. And I think it's also important to keep in mind that the type of disaster we are talking about-- the studies are very carefully done. They have this quasirandom identification, which is great. Hurricanes, tornadoes are hard to predict in their path.

But that also means, how should we interpret the results more generally? We know some areas are much more wildfire-prone, for instance. It's a different type of destruction. And we wouldn't necessarily expect the same interventions to have the same type of results. So I just want to highlight that there is also evidence from other studies that, just in the context of wildfires, people actually start moving out of these areas in response to wildfires.

So there's also-- depending on the disaster we are particularly talking about, there can be also migration responses. There can be responses to the value of the collateral more generally. And in my own work on the response of banks, the large national banks have retrenched to some extent, extending credit to these highexposure areas more, especially to lower-credit-score borrowers, to more riskier borrowers.

And that suggests that here is something where really emergency credit might have the largest oomph because these are the most vulnerable households that otherwise might be rationed out in the credit market. But again, we also, I think, need to think about [AUDIO OUT] on that type of disaster. I'm agreeing with-- it's important that households can maintain the value of collateral.

But again, that is depending on your probability that these disasters happen in your area. We could end up in a state of the world where in certain areas it becomes very hard to insure properties. It's not clear that necessary state interventions to keep people in these areas are necessarily welfare-enhancing, to be quite frank. So I think there needs to be a somewhat broader discussion about targeted interventions and which type of natural disasters we are addressing here. So I'll stop right here.

STACEY: Yeah, I'd love to even jump into obviously, bankruptcy is a very worst-case scenario, and that's such a significant finding of this paper. Maybe if we can just spend a couple of minutes on the impacts of bankruptcy-- obviously, there's the financial impact, but I feel like it goes beyond that, both financially and in other ways. I'd love to just hear a little bit about the impacts that we see from bankruptcy and the power of avoiding it.

BEN KEYS: Yeah. There's a range of costs associated with bankruptcy and lots of studies on this. I think there's of direct administrative costs of running the bankruptcy system, the cost of legal fees, and having bankruptcy courts and bankruptcy judges. There's the large write-offs that lenders are required to do, writing off the debts that are outstanding, and that drives up everyone else's interest rates when those write-offs are large.

So a lender who's trying to stay afloat, if they're losing a bunch of outstanding debt from a bankruptcy resolution, they now need to raise the rates on their existing borrowers. Households respond to the bankruptcy system in a variety of ways, depending, in part, on the chapter in which they file. But in many cases, they are excluded from formal credit markets for quite a number of years thereafter.

Because they're unable to file again for a number of years, they're often not able to accumulate much credit. Afterwards they have to rebuild their credit scores. They're often very difficult to take out a home loan or an auto loan in that subprime credit range.

And there may also be some employment consequences as well, depending on which chapter you file and how you're repaying the bankruptcy over time. So I think it's one of these things where, at the end of the day, we need to have an option for households to resolve their outstanding debts when they get into a difficult situation, but that is a system that creates a number of different burdens.

And so seeing a reduction in the reliance on that system is, I think, along with the other evidence around reducing debts all along the way-- Dan described it as a debt spiral, and I think that's a really apt description that we see at every step of the process from being 30 days behind to 60 days behind. And we can just follow these debts as they fall into distress. I think that is suggestive of a really powerful benefit to these emergency loans.

STACEY: Why do you think this one emergency loan has such a big impact? I have to say, I was pretty blown away by just the size of the impact of just this one bit of money in this one moment. Why is the-- do you have any sense of why the impact is so big?

BEN KEYS: Yeah, I'm happy to hear others speculate on this as well. I think it's because it is very well timed, and it is very large relative to the other kinds of emergency assistance that you might get. So a FEMA grant after a disaster might be a few thousand dollars. This is a program where the average loan is about $25,000, but you can get up to $200,000.

And you're approved up to the uninsured losses. And so if you have a house that's damaged and the insurance adjuster comes and says, well, we'll give you $5,000, but you actually have $50,000 worth of damage, then you have a $45,000 of outstanding damage that needs to be remediated. And this is a program where they will underwrite the loan to exactly that $45,000.

And so the idea is to cover exactly what you need when you need it, and I think that's the key aspect of it. In the paper, we go into trying to understand whether it's because you get to pay it back over 20 years and it's a low interest rate. And so it is subsidized, and there is a-- the federal government does lose money on this program in net present value terms. They have ways of collecting over the long term. But it is kind of a money loser from an NPV sense. It's unlikely that a private entity would offer the same loan and the same terms.

STACEY: Amine and Ralf, I'd love any thoughts you have on this, too, just because that is such a striking finding for me that I just wanted to spend a little time on it because it does seem-- yeah, it does seem like such a big effect, both in the lives of people who are affected by the disasters but also for the larger economy.

RALF MEISENZAHL: So I think, if I may, one of the important things to keep in mind is that there is a significant number of households that do not have any or very little cash reserves at hand to start with. If you go to the Survey of Consumer Finances that the Federal Reserve Board conducts, that is very clear.

And so even small loans or supposedly small loans can help these households immediately. It is very fast. It's very fast redemption, and it helps, exactly as Ben was saying, in the time of needs. And that is, I think, the most important, critical part.

The other part that was also highlighted is if you get this loan to repair your house, you don't have to sell it as a discount because you can't repair it. Then you have to deal with the buyers. They want to have more of a discount and so on and so forth. So that would put you in a really bad situation because you don't have the cash lying around for major repairs. And a lot of households simply don't.

So I think because it is a large, unplanned expense that's not your fault, I think there's a really good case for an additional government insurance that helps to preserve the financial health of these households that come under stress not by their own doing.

AMINE OUAZAD: Yeah. And I wanted to-- when I was reading the paper in detail, I was struck by the fact that the program is administered by SBA, by Small Business Administration, and that's one of the few or maybe the only program of SBA that's providing loans to households.

And I think it speaks to the-- there's some questions of design and implementation that are really key in there, which is, how much time does it take to actually get the funds? How many field offices are there? I saw in the paper-- I think there are three, but how is it implemented? And I think it's worth-- and maybe you won't have the time to talk about this but talking about what's good in the way SBA implements the program.

And the second thing I wanted to say is I think there's a clear parallel between this paper, again, and some of the literature in macroeconomics about the fact that when there's fiscal policy in a time where governments are constrained or liquidity constraints, there's a lot more-- there's a lot higher multiplier than when those governments are not constrained or they don't face a budget constraint.

And I think this is kind of a similar mechanism here, that giving help to households who are not liquidity constraints is going to have a much smaller impact. And perhaps one way to expand the paper would be to get a sense of what are the preexisting conditions of those households.

How much cash do they have saved? That's a very tricky question, actually, from a data standpoint. But that interaction term would be really interesting because it would allow us to target the program towards those households that are most in need. But it looks like SBA is doing a good job here at targeting the program.

STACEY: One of the things I wanted to talk specifically about was flood risk. Obviously, flood risk is growing, and flood insurance has been getting more expensive, especially in certain areas. Oftentimes there are still coverage limitations when a disaster occurs, meaning people will have to find money elsewhere even if they have insurance-- and Ben, you were just talking about this-- and will have to use things like disaster recovery loans.

,p>But what about the insurance companies? Is there a way that they could best manage this issue? And how could flood insurance programs be maybe changed to be more efficient or more helpful?

BEN KEYS: Yeah, I'll take a swing at this first. Most flood insurance in the US is coming from the federal government in the form of the National Flood Insurance Program, and that's a program that ensures--

STACEY: Most of it is now. So most of it's not private. Most of it's government.

BEN KEYS: The vast majority is government, yes. And so they are going to offer you a policy up to $250,000, and so that's the bulk of flood insurance that's available. Lots of homeowners don't understand that flood insurance is not covered by their standard home insurance policy. So they don't realize that that's been carved out of the system and they need a supplemental policy. I think that's a space for consumer education to continue to work harder on.

And so the role that the private market plays in flood insurance is mostly to provide excess coverage above that $250,000. And so if you have a house that's worth $400,000, you might go out and say, well, I'm going to that National Flood Insurance policy, and then I'm going to get a top-up policy, the supplemental policy.

And there is kind of a vibrant private market for that kind of coverage, but the initial issue is really that the take up of flood insurance is just far too low. And I think that people are not getting good information about the cost of flood insurance, about the benefits of flood insurance and getting a real sense of their risk.

I think one of the ways in which people are led astray is these flood maps that we've all seen and we're familiar with, and there's this sort of binary view of it, which is like, if I'm in the flood zone, then my house is at risk. And if I'm outside of the flood zone, then there's no risk.

And hopefully, we've learned some lessons from the last couple of years that even places that on those maps are drawn outside of the flood plain-- and those maps are often very politically drawn boundaries, not topographical or hydrological-- that we should be really aware of these risks. And oftentimes that insurance is quite cheap if you are outside of the flood zone. So I think we need much more continuous awareness of flood risk.

And the overall trend has been a decline in the number of people taking up this national flood risk policy over the last few years, even though the frequency and severity of storms has increased. So I think that's a problem that should be top of the list for policymakers looking at those issues.

AMINE OUAZAD: And jumping in here, I think that something really major that has happened is that when the early literature on natural disasters studied Katrina, 2005, there was actually a decline in debt after Katrina. And that was a moment where there was the highest share, the highest number of policies underwritten. And since then there's been a decline, since Katrina.>

And the reason why I think there's more concern today is because the situation is quite different. Even though there is the availability of NFIP policies and, in fact, NFIP, National Flood Insurance Program, policies are required for households who borrow using GSE mortgage in special flood hazard areas-- I apologize for the acronym-- the take-up is very low. And what that means is that this has much stronger impacts on households' ability to overcome these challenges.

And I just want to say that private insurance is growing very fast, and there are companies like Neptune that provide supplementary insurance. So if we look in terms of the number of policies, it's quite low. In fact, it's below 10%, significantly below 10%. It depends how we measure it.

And the NFIP policies only cover the structure, not the land. That's an interesting aspect because the land is also going to be affected-- we can talk about it at the end of the panel, I think, the value of the land-- and only up to a certain threshold. And after that, we've got to get supplementary insurance.

So that tells us that perhaps there needs to be more information provided. I don't think we have a clear answer to how we should communicate flood risk. There's the problem that it's binary. But how do we communicate flood risk? That's not easy.

RALF MEISENZAHL: And just to add to that, how do how do we communicate flood risk, I think one of the important thing to realize is not being on a coast does not mean you don't have flood risk. And I think that's a common misperception. In fact, when I looked in my own research at our own district around Chicago-- if you go to the north, if you go to Wisconsin, where there's a lot of Lakes, if you get heavy rainfall in these areas, you might well find your basement being flooded.

So I think there is more need, as Ben said, for education on flood insurance, and people need to be more aware in what situation they actually are in. They are now data providers that people can find with their favorite search engine and can basically, for free, look up their house, and I would very much encourage people to do so just to get an idea. That doesn't need to be entirely accurate.

But if you want to learn a little bit more about your own exposure, it might be worth spending the five minutes in the internet to find yourself and see what your exposures actually are. And you might be surprised. When I was looking for a house moving to Chicago, I actually looked for flood risk, and it was not where I suspected it to be.

That's all I'm going to say.

STACEY: Did it affect your home-buying choices?

RALF MEISENZAHL: Yes, yes. You know that Chicago has a-- the Chicago River can affect, actually, neighborhoods further away from the Lake. And I, quite frankly, excluded one or two of these neighborhoods simply because I was like, no, I just don't want to take that risk. Thank you.

STACEY: It's like home-buying with an economist. But Ben, I wanted to touch on something that you mentioned and I feel like has come up a couple of times, but it's a lack of awareness about the policies, not just even a lack of awareness about risk in certain areas, which I think is such a good point, but also just people not understanding that they have to buy these policies. Is that not being disclosed? Is that maybe an area-- that seems like a very solvable problem and maybe cheaply solvable with education. Is that something that could possibly change?

BEN KEYS: Absolutely. And I think it requires a better recognition on the part of the mortgage industry that-- basically, a well-functioning mortgage market depends on having available insurance policies. Insurance bubble-wraps these mortgages so that they can be originated and then bundled and then securitized through the whole process.

And so I think at the outset of that process there should be far more clarity coming from the mortgage lenders, coming from the GSEs about the importance of insurance. And insurance has historically been a kind of a checkthe-box aspect of the home-buying process. It's not something that is top of mind through that process.

And it's also deeply boring. It is just such a boring thing. You're excited to move into your new house. The last thing you want to do is spend an extra day assessing the different insurance options that are out there. But I think people-- and the insurance industry knows this, that people don't understand what is covered by their policies. They don't understand the different limits on coverage. They don't understand what perils have been carved out, like floods or earthquakes.

And so changing the disclosures around this and making it much more transparent and simple to shop around and to understand these aspects-- there's a proposal out from Susan Crawford at the Carnegie Endowment where they recommend making flood insurance opt-out rather than opt-in.

So the idea would be that every mortgage that's originated-- you have to have some form of insurance attached to that policy. A mortgage lender will not accept a loan if there's no insurance on the property. What if the default was that you would put flood insurance on the policy and that you ran all of the numbers and all of the financing with that assuming that the home buyer would take it?

And then if you would like to opt out, you need to sign a form which says, I understand, I am aware of exactly what I am giving up here. And so I like that idea as kind of a baby step that's very behaviorally informed in thinking about how people make these decisions and to underwrite not to where today's insurance prices are but actually underwriting to where they're going to be in the future because the cost of flood insurance is rising and is set to rise on a schedule in many locations.

STACEY: Yeah, yeah. That seems like a great idea, especially considering what a difference it can make with things like 401(k)s and other things. One thing I wanted to make sure to bring up, just because I feel like this has been in the news so much, especially around the fires that we just saw in California and other things-- but what about price caps?

I know that some states, like California, put a limit on how much insurance companies can charge for disaster insurance. I'm wondering how that plays in to the insurance that's offered, to what prices are, and maybe if there are-- how that should be navigated by states and by company.

AMINE OUAZAD: Well, there was an intense conversation in California about the regulation of the insurance premia because there has been a set of regulations. There were reforms in 2024 that changed that. But for a long time, it was not possible for insurance companies to increase premia to reflect the probabilistic risk at a specific location-- the specific risk to be exposed to a wildfire.

And so that introduces frictions. There's work that suggests that the insurance companies that do business in multiple states are using the funds, the earnings they're generating in states where insurance premia are not regulated or are more lightly regulated towards states to subsidize, cross-subsidize their operations in states such as California where insurance is regulated. And one of the things that we've seen-- yeah?

STACEY: People in like states that aren't regulated are paying a premium to supplement it?

AMINE OUAZAD: There's a little bit.

STACEY: Oh. I guess that's kind of how health insurance works in a way.

AMINE OUAZAD: That's true.

STACEY: Anyway, sorry. I didn't mean to interrupt, but that's really--

AMINE OUAZAD: That's true. The purpose of insurance is risk-sharing, so in some sense this is working. But I think what we're seeing is that there is the rise of these plans that are backed by state governments, such as the FAIR plan or Citizens. And that's a sign that the market is not doing well because we need to have a more vibrant system where the states' budgets is not used as a sort of an implicit reinsurance mechanism if I can say it like that. And so flexibility is important.

When we control prices, what happens is that quantities decline. As the economist, I think that's one of the first things that we teach is if we control prices, then we get shortages. But I think the conversation shouldn't stop there because there's a question of affordability as well. And if insurance premia are unaffordable, that sends the signal to households that they are facing a much higher risk.

And I just want to make one recommendation that's actually very, very, very simple but I think would change a lot of things as we were talking about floods. And I would like to phase out this expression of the 100-year floodplain. The 100-year floodplain is not a 100-year floodplain because it suggests that the flood will happen every 100 years, once a century.

That's absolutely not true because it's a simple math exercise that we do in high school that if there's a 1% probability of a flood every year, that means that we have something like a 36% probability of being hit once during the life of a 30-year mortgage. So that's considerable. If we were communicating that probability, no, 36%-- that becomes much more tangible than 1%.

We've all studied math and statistics and all that, but 1%? I don't what that means. But 36% I think I know. I know that that's considerable. And I think for wildfires we also need to communicate things in a more tangible way so that we understand the risk. Yeah. [INAUDIBLE]

RALF MEISENZAHL: If I may--

STACEY: [INAUDIBLE]

RALF MEISENZAHL: So yes, there are a lot of different-- we have state regulation for insurance, so it really is state-dependent, what we are going to see. But there are limits to cross-subsidization. At the end of the day, in the long run, you have to be able to basically make a zero profit on the contract. That's kind of the goal.

And this is why places might become completely uninsurable, or house prices just have to drop, one of the two. And it's like-- because subsidies only run for so long. It's true. California has started to address this. And California in particular, the FAIR Plan has much more take-up, Citizens in Florida.

They tried to reprivatize these insurances but with up to 25% price increases. So these are very substantial increases that the consumers are going to face. And in the end, the larger conversation about risks coming from catastrophes, from these events, insurance prices, rebuilding prices, and house prices and where people will be able to afford to live-- there are places where people get literally priced out now, or they have to go without insurance.

That will have, in the long run, very complicated consequences. So if only the affluent can afford living in certain areas because the lower-income households get priced out-- I would draw the comparison a little bit to what happened once in Silicon Valley when you couldn't get a gardener to come unless you woke up at 6:00 AM on a Sunday because there were simply not enough workers who could afford living in these areas.

And there's a certain risk that that is going to happen as well, that you see an extreme segregation who can live in certain areas and who cannot. And I honestly don't have a really good policy prescription for that, even though I completely support that-- don't call it a 100-year floodplain. That I'm fully behind.

STACEY: Well, does the existence of these programs, of the recovery loan programs-- does that contribute to or cause people to maybe take up insurance at a lower rate? Do people maybe think that they don't have to get insurance so much? Is that one of the impacts that we see?

BEN KEYS: So I'm skeptical that that's what's driving the lack of take-up of insurance. I think it's more directly related to a lack of understanding about risk and then the rising cost of insurance. And so part of why we've seen flood insurance decline is that there has been an effort to more closely align flood risk with the actual price of the policies. It was massively mispriced previously.

If you think about what's going on in the federal disaster loan program that we're looking at, about 50% of all applications are rejected. And so about half of the people who apply for these loans do not get them, and most of the people who are applying don't understand necessarily what are the determinants of-- how do I qualify for this loan?

A high credit score helps a lot. A low debt-to-income ratio, suggesting that you can make those monthly payments, helps a lot. But I don't think a lot of applicants are going to be thinking about that beforehand, before this disaster. So this idea that the federal government is going to step in and make you whole is absolutely a myth.

I think anyone who's lived through a disaster-- and I have family who have. I think that that's-- that kind of damage to your property is really scarring. It affects you quite deeply. And I think that there's a sense in which even the people-- I think one of the other takeaways from our paper is that even the people who receive these loans are worse off after the disaster.

And so they do have more elevated bankruptcy rates. They do have more elevated debt in collections. They do have other indications of financial distress relative to a control group of people outside of the disaster area that we used as a comparison group. So even among the group that's receiving these loans, they're still worse off on net.

And so I think putting all that together suggests that this isn't a very tight safety net that households should rely on and say, I'm going to drop my insurance policy because I know that FEMA and the Small Business Administration have my back. That's just not a good approach to thinking about risk management at the household level.

STACEY: I'm really surprised that half of the people who apply aren't able to get a loan. That's a high number. That's a lot of people. What are the financial implications for them? What do we see for people who aren't able to get a loan?

BEN KEYS: Yeah, that's exactly the comparison group that we're using in the analysis. And so there's elevated bankruptcy rate, more debt in collections, more financial distress, less ability to replace their cars. The group that doesn't receive these loans, based on our data is really excluded from traditional financial markets. They are not finding other sources of credit to help them through this process. And so in that way, the government is providing this critical role of liquidity infusion at exactly the right time.

STACEY: Are there policies that could be put in place to help people who are denied a loan or who, for whatever reason, are unable to access that loan? Or would it just be expanding this program? Or would it be-- are there maybe other policies that we could think about that could help those families?

BEN KEYS: I'm happy to hear what others think. I think it's both, ex ante. It's a before-the-disaster problem and an after-thedisaster problem. I think some of these things we're talking about about improving insurance coverage, making people aware of their risks and making sure that they are taking on the insurance policies that they need-- I think that's probably the first step.

And then in terms of expanding this program, I think, given our findings, that's certainly suggestive that expanding the parameters a little bit in every direction to bring more people into this program would be beneficial. And exactly how you design that is one of the interesting things that we've played around with a bit with our data and thinking about different groups of people.

The loan program has shifted over the years in terms of who they're more willing to lend to and less willing to lend to on the basis of repayment patterns and things like that. So they're very aware of this, the differences across how they underwrite the loans.

And so yeah, I do think expanding a program like this or expanding some of the grant programs that Dan talks about in his tornadoes paper would be another one. Those grants, in the event of the aftermath of a tornado, looked quite beneficial to households, and so it looks like we should be doing more of that, from a government standpoint, to help people through these difficult situations.

AMINE OUAZAD: And perhaps jumping in here is the fact that we may want to talk about what we can do before an event happens, before a natural disaster hits. And it would be interesting to see if the funds are used for further adaptation to repeated natural disasters. For wildfires, there are simple measures that households can take, such as cutting trees, clearing the land around the home so that there are no embers that can ignite the house from inside, from--

And so there are firms like IBHS that help households fortify homes. So they help in designing better homes. And we've seen in the recent Los Angeles fires that some homes have survived these wildfires, and so even though they were within the wildfire perimeter that was delineated by the firefighting departments, those homes were left practically untouched.

And we know that there are some good practices. There are also ways, from satellite imagery, to see what's the kind of tree canopy that there is around the house at high resolution to be able to say, this house is likely to burn down during a wildfire event. And so I think what we need to understand as economists is how we can think about multiple events, multiple natural disasters that affect a home repeatedly and whether these lines of credit have a different impact for a second natural disaster, a third one, and so forth as we know now that this is a recurring event.

RALF MEISENZAHL: I will just add because unfortunately I will have to leave-- just the last word for me is I think, yes, talking more about prevention and preventive measures is something that should be part of that conversation that's actually also where insurance companies, private insurance companies, could be quite helpful by pricing these discounts accordingly. And that might provide a lot of incentives. So thank you very much.

STACEY: Thank you, Ralf. It was wonderful having you.

RALF Thank you.

MEISENZAHL:

STACEY: Should we keep going, Kristen? Are you jumping in because of time?

KRISTEN BROADY: --in because of time, but I would like to thank you, Stacy. It is always wonderful to have you moderate our events. I'd also like to thank our panelists, Ben Keys, Ralf Meisenzahl, and Amine Ouazad, for such an enriching and enlightening discussion. I learned a lot about credit and how they are able to impact the financial position of households following a natural disaster. I hope that you learned as much as I did or more.

So thank you to Dan Hartley, as always, for presenting his wonderful research. We will be sending a postevent survey, so please be on the lookout for it. We value your feedback, and your feedback helps us to develop more events like this one.

A recording of today's event and a summary will be available on our website. That's chicagofed.org/mobility, and so that will be available in the next few days. Again, thank you for joining us, and have a great afternoon.