How Do Natural Disasters Affect Students?, with Isaac Opper

Notable Quotes

• Natural disasters have long-term consequences on education outcomes: “We actually thought there was a pretty good chance that, at least for test scores, we’d see this temporary blip, and then it’d go back to what it was before. We actually see no evidence that things are temporary and that test scores or graduation rates or college attendance increase two or three years afterward. We can only look about five years post-disaster based on our data, but we see no real evidence that things improve over that course.” (10:57)
• The cost of natural disasters to human capital can exceed the cost to physical property: “If a disaster causes $1,000 in per-person property damage, we find that, to a rough approximation, it also causes about $1,000 in human-capital costs per person … That especially hits the students who are in elementary school, high school, and college. If you want to focus especially on these students and young adults, the per-person cost or per-student cost is actually quite a bit higher than the average property damage cost.” (15:57)
• Climate change may exacerbate the risk of learning losses due to disasters: “My understanding … is that most of the evidence suggests that these disasters are going to increase in both frequency and size … We have to figure out how to mitigate some of these responses—maybe put things in place so that schools and students are more resilient … If we don’t have procedures in place or figure out how to mitigate this, that’s going to be yet another channel through which climate change is going to affect our lives and … affect economic output, as well.” (24:36)

Top of the Stack

• “The Effect of Natural Disasters on Human Capital in the United States” by Isaac M. Opper, R. Jisung Park, and Lucas Husted
• The Invention of Nature: Alexander von Humboldt’s New World by Andrea Wulf

The Full Transcript

Daniel Raimi: Hello, and welcome to Resources Radio, a weekly podcast from Resources for the Future. I'm your host, Daniel Raimi. Today, we talk with Dr. Isaac Opper, an economist at the Rand Corporation and a professor of policy analysis at the Pardee Rand Graduate School.

Along with a couple of coauthors, Isaac recently published a paper called “The Effect of Natural Disasters on Human Capital in the United States,” which quantifies how natural disasters affect students' learning outcomes and their ability to earn income over the long term. When we think about the effects of natural disasters, we often think first about the damage to physical infrastructure and risks to human lives, but these events can upend people's lives in all sorts of other ways that can be just as—if not more—impactful than the physical damages that are easier to see and measure. I'll ask Isaac to help us understand this issue in today's episode. Stay with us.

Isaac Opper is joining us from the Rand Corporation in sunny, beautiful Santa Monica, California. You were just making me and our producer, Elizabeth, jealous about the wonderful weather you're experiencing out there. Thank you for joining us on Resources Radio.

Isaac Opper: Of course. Thanks so much for having me.

Daniel Raimi: So, Isaac, we're going to talk about a really interesting paper that you recently published with coauthors on the effects of natural disasters on human capital.

Before we do that, we always ask our guests how they got interested in topics related to the environment—whether that interest started when you were a kid or whether it developed later in your life. So, what drew you into this field?

Isaac Opper: I think my interest goes way, way back and probably comes a lot from my grandfather, who was a forest ranger in Montana, where I grew up. So, in some sense, my love of nature really came from the walks that I took with him as a kid. That interest really gained steam when I was about 10, and I read in the local paper that the fish in the creek that ran through our little town of Lewistown, Montana, were contaminated with a chemical. So, you couldn't eat the fish, because it was cancerous.

Naively, I decided that my fourth-grade science project was going to be to find the source of this pollution. We ended up taking a bunch of soil samples in the creek and used a lab to test them for this chemical called polychlorinated biphenyl, or PCB, and we actually managed to isolate it to a relatively small stretch of the stream. From that moment on, I was utterly convinced that I was going to be an environmental scientist when I grew up.

Fast-forward 25 years, and, of course, now, I'm not an environmental scientist. I'm an economist. Not even really an environmental economist, but more a labor-and-education person. So, life doesn't necessarily unfold the way you think it will as a fourth grader. But the interest in environmental issues has really been there for as long as I can remember.

Daniel Raimi: That's so fascinating. Did they end up identifying the actual location of the PCBs, and was there enforcement? What was the end of that story?

Isaac Opper: Yeah, there was a huge cleanup run by the US Environmental Protection Agency and some of the local state agencies, as well. I don't remember the exact timing, but I think anywhere from 5 to 10 years ago, I think the fish are now officially deemed clean and able to eat. So, it's a really wonderful end to the story.

Daniel Raimi: Wow, that is really fascinating. So, we could talk about fish in Montana for a long time, I suspect, and that would be fascinating. But instead, we're going to talk about this recent paper that you wrote with coauthors. Again, the title of the paper is “The Effects of Natural Disasters on Human Capital in the United States.”

Let's start with a couple of basics. I think when most of us think about natural disasters, we probably think about the effects that those disasters have on property and maybe the environment—if we think about wildfires, maybe human health. But what is human capital, and how, just in a general sense, might human capital be affected by different types of natural disasters?

Isaac Opper: I think that's a really good question, and a good first question in some sense, because people, myself included, confusingly use the term human capital in a lot of different ways. One example is, if you hear about human capital, it's often sometimes used as a way to capture components of an individual's well-being that don't show up in the official economic statistics, like the country's GDP. That's actually not really how we're using the sense of human capital.

We use a more narrow definition that, I think, is best stated by this wonderful economist, and actually the most recent Nobel Prize winner, named Claudia Golden, who, in this definition, says that human capital is just the stock of skills that the labor force possesses.

So, we're especially going to focus in this paper on education, which you could think of as the production of human capital. We know, as I said earlier, that I think I'm mostly an education economist, and so I've read lots of papers that have talked about how this education process can be affected when either students or teachers experience this kind of large disruption in their lives.

Of course, natural disasters are a huge disruption. That was really the motivation to start looking at this—thinking that it's worthwhile to study directly what the effects of this big disruption caused by natural disasters have on the education process.

Daniel Raimi: That's great. This might be intuitive to people, but can you just walk us through an example of the chain of events that leads a natural disaster to affect someone's educational attainment?

Isaac Opper: That's a fascinating question. Frustratingly, there's a lot of chain of events that could happen. One thing that I will just say from the beginning is that this paper is looking at the overall effect, not the particular mechanisms. But you could imagine that somebody is in community college, and a natural disaster causes $2,000 of damage to their car. So, they have to take a second job out and that causes them to drop out of college. That's one plausible story. It could be just a simple story of missing school and canceled school for a while for a fifth-grader, and that causes the reduction in learning. There's a lot of different pathways. As I said, I think we could study a lot more, and hopefully, people will study a lot more about which of these pathways is driving the results. We're really looking at the big picture of overall what are the effects.

Daniel Raimi: That's great. I think that just having a couple examples helps us—or at least helps me—think about the intuition here. So, I want to ask you a question about your data, briefly. We're not going to spend a ton of time on data, but can you just give us a flavor of the data you're using and how it allows you to try to answer this question of how disasters affect people's educational attainment?

Isaac Opper: We didn't collect any data ourselves for this paper but really piggybacked on other researchers and used what's called “secondary” data, or data that others have collected. We basically need two types of data—one data set on natural disasters and one data set on education outcomes.

For the disasters, we use a data set that consists of all presidential disaster declarations, which are these formal declarations that have occurred about over the last 20-odd years. There's actually a lot of these declarations—about 400 counties experience one of these every year. So, we can be pretty confident that that captures, I think, the majority of the sizable disasters in the United States. Very helpfully, there've been other researchers that have used this data set, cleaned it up a bit, and provided estimates of how much property damage each of these disasters have caused. That's really the data set we're using to measure disasters.

On the education side—if you're a parent listening, you probably already know this—students often take tests at the end of each year. So, we use these test scores for students in grades three through eight. We use high school graduation measures and college attendance. All of these bits of data sources end up getting reported to the US Department of Education. Then, there are relatively well-known data sets that aggregate these components, so we're not compromising students' confidentiality. There've been other researchers that have done a really good job of cleaning these and making them available to researchers. So, we're using these data sets to measure education outcomes.

Daniel Raimi: That's great. That makes a ton of sense.

Let's just jump over the methods, which are really interesting and would be great for a more technical podcast. But we'll let people check out the paper to dig deep on the methods there and jump over to the results. Can you just talk us through some of your high-level results in terms of how natural disasters affect the learning outcomes that you're looking at in these data?

Isaac Opper: Definitely. So, you said we're jumping past the methods, but I'll say they're pretty straightforward in the sense that all we do is look at what happens after a disaster relative to the year before the disaster. And so, as an example, we find that students’ test scores in areas that were hit with a disaster tend to be lower in the year after a disaster relative to the year before the disaster. We also find evidence that it hurt high school graduation in those areas—not of students who were seniors at the time the disaster hit and who were almost ready to graduate, but actually the students who were juniors at the time the disaster hit. Then we also find, as I mentioned, pretty strong evidence that disasters also reduce college attendance in these areas.

Daniel Raimi: That's really interesting. You say that you're measuring these events before and after the disaster, but you also talk a little bit about the persistence of the effects—how long they last. I'm curious if you could talk a little bit about what you find in terms of how well people are able to recover after they have experienced a natural disaster like this, or if the effects persist over a long period of time.

Isaac Opper: Just as we could look at what happens in the year after a disaster relative to the year before, we could look at what happens two years after a disaster or three years after a disaster. We actually thought there was a pretty good chance that, at least for test scores, we'd see this temporary blip, and then it'd go back to what it was before. We actually see no evidence that things are temporary and that test scores or graduation rates or college attendance increase two or three years afterwards. We can only look about five years post- disasters just based on our data, but we see no real evidence that things improve over that course. So, it does seem to have pretty persistent effects.

Daniel Raimi: That makes me wonder about COVID, actually. I know, in some ways, we're still in the COVID era. In other ways, maybe we're a little bit beyond it. But I'm curious if these results that you're finding—if they're consistent with what you might have seen in the literature. Again, this isn't my field, so I don't know if people have been doing studies on this. But I'm wondering if what you're finding might be consistent with the impacts to learning outcomes that students might have experienced in the aftermath of the lockdowns with COVID.

Isaac Opper: Again, that is a really good question. I haven't done any research on this, but I know some of my Rand colleagues have. COVID is, obviously, a very different type of disaster than the one we're looking at, but I think you could certainly consider it a type of natural disaster.

There's lots of evidence that COVID harmed learning outcomes across the board. I think one thing that we can learn from that research that we don't talk about, but that I think is a really important part, is one of the big findings from the literature on how people were affected by the pandemic: There's lots of different disparities, there's lots of equity issues, and it's really the most needy students who were the ones that were affected the most. I think one of the real limitations of our paper is that, just based on what data we have, we don't get into these kinds of equity considerations much at all. But I think, especially from COVID, that's a really important question, because COVID did have these huge disparities.

Daniel Raimi: Is that something that you're actively following up on in this work, or that others might be following up on?

Isaac Opper: It's something that I hope to follow up on. So if you're an Institute of Education Sciences (IES) grant reviewer out there, Daniel thinks you should fund us. Rand is a soft-money organization, so we need to get funding to do our research. So, we actually recently put in a big grant to try to look at these equity issues and think through other data sets that could allow us to answer that a little better.

Daniel Raimi: That's fascinating. Well, you've got the Resources Radio stamp of approval on your grant application.

Isaac Opper: Perfect. I'm sure that's exactly what we need.

Daniel Raimi: That's what the reviewers are looking for right there.

Well, let me ask you one more question about your paper, which is trying to compare in some ways the education outcomes that you are measuring with other outcomes from natural disasters that we might be more familiar with, like billions of dollars in property damage or, I don't know—maybe lives lost from a hurricane or a wildfire. How would you put the magnitude of the effects you are seeing here in context with some of those other metrics we may be more familiar with?

Isaac Opper: Again—a really, really good question. First, I'll just say that, even as an economist, I think I would argue that there's no real way to perfectly compare the roof getting ripped off someone's house to somebody dropping out of college after a storm or having their test score reduced a bit. So, we're not trying to put this precise dollar value on everything, because that is what we do as economists. I think, instead, the real goal is just to try to get a sense of scale. Are these human-capital costs that we see big relative to the things that we can more easily measure and are more easily visible, like property damage, or are they relatively small compared to these physical damages? We try to get at that by looking at what's called the net present value of earnings, where we're really trying to approximate the earnings loss due to reductions in the education outcomes that we see.

The thing we like about this measure is that it's comparable to the property damage in the sense that if somebody has $500 in property damage, in theory, you'd have to write them a $500 check to make up for that damage. Similarly, if we say there's a $500 reduction in human capital, in theory, you'd have to write them a $500 check to make up for that loss. When we do that, when we try to quantify it in that way, we actually find that the magnitude of the damages to human capital are, I would say, roughly on par with the damages to physical capital.

If a disaster causes $1,000 in per-person property damage, we find that, to a rough approximation, it also causes about $1,000 in human-capital costs per person. I think one thing I want to note about that is property damage costs aren't the same across everybody in the county. Some people might have $1,000; some people might have $200 in property damage. In human capital costs, that's especially true. In particular, this $1,000 cost per person that I mentioned—for a $1,000 property damage disaster, that especially hits the students who are in elementary school, high school, and college. If you want to focus especially on these students and young adults, the per-person cost or per-student cost is actually quite a bit higher than the average property damage cost.

Daniel Raimi: That's really interesting and that makes a lot of sense.

So, I'd love to ask you now to turn to policy implications. This, again, is totally outside the stuff that I study. I'm curious if you could just give us a lay of the land of what some of the possible policy responses might be, and what you're hoping policymakers are able to draw from your key findings.

Isaac Opper: Rand is all about policy implications. It's something that I think a lot about in both this research and other research. I think a, frankly, frustrating part of what I've done so far is that I don't really know what to do about it.

We've talked a little bit before. I think tons of money goes into rebuilding the physical capital after a disaster, and very little goes into explicitly rebuilding the human capital. In the long run, a natural policy implication is that we should equalize those a little bit and provide more funding for rebuilding the human capital. But, in the short run, I would say, without a great sense of the mechanisms behind this reduction—as I said there's lots of different ways it might be affecting things—without a great sense of what's causing these, I don't think we know exactly how to rebuild human capital in the same way that we know how to rebuild a highway or a hospital. I just don't think that's as obvious about how to rebuild human capital.

You mentioned COVID. I think COVID is a good example of that. There's a lot of money trying to improve education outcomes, and there's a lot of good work. I think it's helping to some extent, but we don't have a perfect answer of how to do it. It's obvious that the policy implication right now is just to spend a bunch more money on rebuilding the human capital. I think—again, this is maybe a plug for the IES reviewers on the grant—I think the hope is to better understand what's causing these effects, and then we can maybe think about how to remedy those.

Daniel Raimi: Yeah, that's really interesting.

And I forgot to ask you when you first mentioned it—I don't actually know that acronym IES. What does that stand for?

Isaac Opper: Sorry. You told me not to use acronyms, and of course, I did.

Daniel Raimi: No worries.

Isaac Opper: That is the grant that the Department of Education runs for researchers.

Daniel Raimi: Okay, cool.

You just alluded to this, but I'm curious if you can talk a little bit now about what are the tools in the toolbox that exist that policymakers might consider if they observe this learning loss and they want to try to make up for it in some way or compensate for it in some way. Can you just talk through some of the efforts that exist? What are the programs that people are trying out? What are the lessons that are being learned, if any, to this date? And how do you think they might be applied in the context of recovering from disasters?

Of course, I want to preface your answer by saying that, obviously, you're looking to do more research on this to have more empirically grounded answers to this type of question, but I'm wondering if you can just give us a little bit of examples and intuition.

Isaac Opper: I hate to keep coming back to more research, but that is something else actually that we're trying to get funding to do a little bit. Anecdotally, it seems like there are programs that are in place at various districts to try to address what happens if there's a disaster—what their process is for making sure students still get educated and potentially moving to an online program. One of the things that is both good and bad about the US education system is it is very decentralized. So, different districts are doing different things; potentially, even different schools are doing different things.

That's actually one of the other things on our research agenda—to try to document that a little bit more systematically to get a sense of what is out there now. Ideally, we can then combine that with some of the empirical work to say, "This is what seems like is working, this is what seems like is not working." So, it is very much something I'm interested in, but again, frustratingly, I don't think I know the answer right now.

Daniel Raimi: That's interesting. I'm curious, even if you can just tell us, what would be an example of a program that a school district might try to do or that a state might try to do to make up for these challenges? Would it be like summer school; you mentioned online learning? What are the actual activities that would be involved?

Isaac Opper: Just anecdotally speaking, I think they're the exact things that you mentioned and things that make sense that, if it's just reducing the number of days of school, because everybody has to stay home for a while and that's a two or three-day thing, maybe doing summer school or adding onto the school year if it's a longer term. It seems like most districts have in place something where they can shift to online learning if the school building itself has some damages done. So, summer school, after school programs, extending the school year, and the ability to get computers to students—we went through a lot of that with COVID of trying to shift to online—I think a lot of those are also in place for more types of disasters like hurricanes or floods.

I'll also say, as far as I can tell, the college side of things is even, if anything, more decentralized, where individual colleges are doing different things. And I don't have a great sense of what colleges are doing at all now. Say a community college student, again, has a financial hit because of a disaster and might end up taking a second job, and that causes them to drop out. I'm not really aware of anything that's happening there. That's not to say that nothing is happening; I just don't know of it. I think that's another thing that I think we should look at, because a lot of the costs really do seem to come from the fact that people don't go to college at the rates they did without the disaster happening.

Daniel Raimi: That makes sense.

One last question before we go to our Top of the Stack segment is about climate change. I'm sure people have been thinking about this during our conversation. I'm not a climate scientist, but my understanding is that the theory tells us that as climate change worsens, certain natural disasters like tropical storms, wildfires, and other extreme events are likely to become more intense and likely more damaging. I think there's actually some debate about this—about whether the empirical record to date tells us or gives us the climate signal in natural disasters—but we don't need to get into that debate. I'm curious about how you think about climate change and the results that you're finding in this paper. What are some of the things that you think about when you think about a changing climate and how they might exacerbate, or not, the educational outcomes that you and your coauthors look at?

Isaac Opper: So, like you, I'm not a climate scientist and, like you, I've been a bit confused when reading the literature, because you see both sides a little bit. That said, my understanding, like you, is that most of the evidence suggests that these disasters are going to increase in both frequency and size. So, my read of that literature and our paper, I think if anything, really just emphasizes the fact that we have to get this right, we have to figure out how to mitigate some of these responses—maybe put things in place so that schools and students are more resilient. Again, my understanding of the literature suggests that this is just going to keep happening more and more and more. If we don't have procedures in place or figure out how to mitigate this, that's just going to be yet another channel through which climate change is going to really affect our lives and reduce—even if you're just focused on economic output—another channel through which it's going to affect economic output, as well.

Daniel Raimi: Yeah, for sure.

It would be fascinating to talk about this in the context of the social cost of carbon, which many of my colleagues at Resources for the Future (RFF) work on in great depth. You could imagine trying to incorporate this sort of thing into the social cost of carbon, which at the moment, I don't believe it is.

One of the other things that I take from your paper is that even setting aside climate change, let's imagine these natural disasters don't get any worse. It's still a really big deal and still incredibly disruptive and impactful on people's lives and the economy. I'm curious what you think about that.

Isaac Opper: I think that's exactly right. I agree that I've thought a little bit about the social cost of carbon and thinking about this in that context. There's also some evidence, I will say, that disasters affect long-term growth rates. So, education is one channel through which it could be affecting these long-term growth rates. Again, to the extent that climate change is going to affect disasters, that's another drag on long-term growth, which is something, I think, we all should be a little concerned about.

Daniel Raimi: Yeah. For sure.

Well, Isaac, this has been a really fascinating conversation. I've been out of my depth for most of it, so I appreciate you keeping things straightforward so that myself and our listeners can really understand the takeaways here.

I'd love to ask you now to recommend something that you think is really great. It can be related to the environment or not—we’re not all that picky. So, what's at the top of your literal or your metaphorical reading stack?

Isaac Opper: Yeah. A book I just read that I really, really liked. It was called The Invention of Nature by Andrea Wulf—I think it was a bestseller in 2014 or 2015. It's about Alexander von Humboldt, who was this adventure scientist in the late eighteenth century, early nineteenth century. For one, it's just a great story, learning about all of his adventures and findings, but I think it's also a thought-provoking one, in the sense that—we started talking about my love of nature, and that begs the question of, What really is nature? We mentioned I grew up in Montana and now live in Los Angeles, so that's an especially pertinent question, because nature is a little bit more subtle here, I think, than it is in Montana. So, I really both enjoy just the story of that book, but also thinking through how we think about nature and going back to one of the first people who really thought about that in the modern conception.

Daniel Raimi: Wow, very cool. I have not heard about that book. That looks really, really interesting. That's a great recommendation and, again, a really fascinating paper. Thank you so much for writing it and for sharing it and for coming onto the show and helping our listeners understand it. We really appreciate it, Isaac Opper from the Rand Corporation.

Isaac Opper: Thanks so much for having me.

Daniel Raimi: You’ve been listening to Resources Radio, a podcast from Resources for the Future, or RFF. If you have a minute, we’d really appreciate you leaving us a rating or comment on your podcast platform of choice. Also, feel free to send us your suggestions for future episodes.

This podcast is made possible with the generous financial support of our listeners. You can help us continue producing these kinds of discussions on the topics that you care about by making a donation to Resources for the Future online at rff.org/donate.

RFF is an independent, nonprofit research institution in Washington, DC. Our mission is to improve our environmental, energy, and natural resource decisions through impartial economic research and policy engagement. The views expressed on this podcast are solely those of the podcast guests and may differ from those of RFF experts, its officers, or its directors. RFF does not take positions on specific legislative proposals.

Resources Radio is produced by Elizabeth Wason, with music by me, Daniel Raimi. Join us next week for another episode.