Pollution Causes Crashes: Particulate Matter and Traffic Accidents, with Travis Roach

Notable quotes

• Exposure to air pollution negatively affects cognitive functioning and mental health: “One of the most famous examples is probably the case of umpires … On days in cities where there’s especially high PM2.5, we see that these umpires make more mistakes … A lot of folks have looked at performance on standardized tests after there’s exposure. We’re also seeing this with behavior and depression. In more of the medical and epidemiological literature, we see that there are later-in-life outcomes that are very bad.” (6:49)
• Air pollution may increase the incidence of fatal car crashes: “What we see is that, by having more pollution exposure, there’s a 1 percent increase in fatal accidents. If there’s 35,000 accidents per year, then what we’re finding is that about 350 accidents could be due to a single unit change in PM2.5.” (10:35)
• Clean Air Act has improved air quality for environmental justice communities: “The housing around our most polluting areas tends to be cheaper … If there’s housing that is cheaper, that tends to be lower-income folks who are living around that area, so they’re specifically burdened by a lot of the environmental pollution we’ve seen thus far in our country’s history. But … the Clean Air Act and its amendments have really reduced the burden on, specifically, lower-socioeconomic folks who happen to be living in these areas that have high industrial activity.” (21:00)

Top of the Stack

• “Negative Externalities of Temporary Reductions in Cognition: Evidence from Particulate Matter Pollution and Fatal Car Crashes” by Anne M. Burton and Travis Roach
• PurpleAir sensors and maps that measure air-quality data
• Prairie Up: An Introduction to Natural Garden Design by Benjamin Vogt

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 Travis Roach, an associate professor of economics and director of the Central Policy Institute at the University of Central Oklahoma.

Along with coauthor Anne M. Burton, Travis recently published a working paper that quantifies how airborne particles increase fatal traffic accidents in the United States. This finding builds on a fast-growing stream of research that demonstrates how harmful fine particulates are both in the short-term and the long term for human health and decisionmaking. Travis will describe what his research finds, how it fits into this growing body of evidence, and what it all means for public policy, environmental justice, and more. Stay with us.

All right. Travis Roach from the University of Central Oklahoma, welcome to Resources Radio.

Travis Roach: Thanks for letting me be here.

Daniel Raimi: It's our pleasure. I'm really excited to talk about this new working paper on a particular matter in traffic accidents.

But before we get into the substance, we always ask our guests how they got interested in working on environmental issues, either as a kid or later in life. What drew you into this line of work?

Travis Roach: It's probably a lot of things. I should give credit to two fantastic public school teachers I had in high school, Penny Smeltzer and Bill Blaine, who taught me advanced placement statistics and economics. They're truly master teachers. I didn't know it at the time, but they really set up what would be just like a crucial foundation for the rest of my life.

Going on into college, I went up to University of North Texas, and I was going to be a music major there, but then I stayed in economics, partly because I had this great background. I was in my environmental economics class, and it was just so cool to go one week talking about preserving wildlife in Africa and saving elephants to then—this is in Denton, when the fracking ban started. Denton County or the city of Denton passed a ban on fracking, and after the fact, the state of Texas came in and said, "No, you can't do that. We don't want that form of limited government."

Then, later on, I go on to graduate school at Texas Tech University, and I basically had front row seats to the wind boom. Every time I'd go visit family, I'd be able to see a new wind farm pop up on my way to or fro. It's one of those cool little endogeneities where, Am I interested in energy and environmental issues because of what was around me? Or was I interested in it, because I would go on to be energy and environmental economist? It's probably a little bit of both, but it's been this lifetime of having this surrounding me and this cool policy space and interesting things happening.

Then, I move up to Oklahoma City, where I'm at University of Central Oklahoma, and it's one of the capitals of the fracking boom. Energy has kind of been a part of my life ever since I went off to college.

Daniel Raimi: That's really interesting. Where in Texas did you grow up?

Travis Roach: I grew up in Austin, Texas.

Daniel Raimi: Excellent. Great city, Austin, Texas, of course.

Travis Roach: Yeah.

Daniel Raimi: All right. Well, I’m sure we could talk about fracking and wind for the entire episode, but we're going to talk about something else, which is your recent paper with Anne M. Burton. It's a working paper. We'll have a link to it in the show notes. The paper's called “Negative Externalities of Temporary Reductions in Cognition: Evidence from Particulate Matter Pollution and Fatal Car Crashes.”

Let's just start off by defining “fine particulate matter.” Most listeners probably know this, so we don't have to spend too long on it, but we also call this PM2.5. Can you define PM2.5 and then help us or remind us what its main sources are?

Travis Roach: Sure. PM2.5—particulate matter is measured in microns, and the 2.5 refers to the literal size of this itty-bitty little particle. One way to think about this is if you grabbed a strand of hair on your head, that probably has about five PM10s in there, so it’s 10 microns wide. Then, each one of those little 10 microns, we're going to split up four times. A single human hair has, 20 or 30 PM2.5s within it. It's an incredibly small particle.

The issue here is that it is a result of basic combustion processes. What's on the top of my mind is the Smokehouse Creek Fire that's happening in West Texas, but we've seen these wildfires happening up in the Northwest and these plumes of smoke that are going across the country. The West Texas fire right now … there were reports of Canadian folks seeing the smoke that was happening down there. It travels a very great distance. When we think of wildfires, that's one of the main sources of PM2.5. But the more impactful ones—the ones that hit us on a day-to-day basis—come from coal plants, natural gas plants—basically any thermal unit—and then, of course, tailpipe emissions.

Daniel Raimi: Right. If I'm not mistaken, it's mostly diesel, more than gasoline, that produces PM2.5. Is that right?

Travis Roach: Diesel has a higher concentration per cubic meter, but the way we measure this is in micrograms per cubic meter. Diesel is very dense.

Daniel Raimi: Let's do a quick literature review. There have been a decent number of studies in the last few years that look at PM2.5 and try to estimate how it affects human health and also cognitive function. Can you give us some highlights in that literature?

Travis Roach: It does feel like a little bit of a gold rush of folks studying this topic, because it is just shown in so many different contexts and countries to be this really impactful and harmful source of pollution. Some of the very early work looked at animal studies and was looking at mammalian behavior and seeing that orangutans and rats (the traditional medical study participants, if you can call them that) were reacting differently when they were exposed to high pollution.

Then, this has left the mammal route and come into the more human route, where we can see how cognitive function among people is changing. We've seen this with cognitive tasks. Various studies will give you a slate of questions and ask you to do this or that, and we can measure your exposed pollution depending on what city you live in. Or in some cases, they're literally exposed to pollution in situ. They're putting that right there.

We're seeing that cognitive performance declines. One of the most famous examples is probably the case of umpires, which is one of those great examples of seeing something that's very verifiable. We can draw a box and see what's a strike and what's not a strike. On days in cities where there's especially high PM2.5, we see that these umpires make more mistakes. This has gone into various different education outcomes. A lot of folks have looked to see how performance on standardized tests go after there's exposure. We're also seeing this with behavior and depression.

In more of the medical and epidemiological literature, we see that there are later-in-life outcomes that are very bad. The keystone paper here, maybe, is from Tatyana Deryugina and Julian Reif and their coauthors, who see that there are bad effects on morbidity due to more early life exposure to PM2.5. Then, Claudia Persico has—I'm going to say fantastic; it's a very sad paper—showing that there's a direct link between PM2.5 exposure and suicide. Then, if we can keep going this route of seeing how our brain process is affected and how might that spill out into regular economic behavior, we're also seeing that crime is affected.

There's a few gray papers on this, but one is kind of using the proximity to a highway and whether the wind is blowing upstream or downstream of that highway. It's this really neat natural experiment where you can see, Does crime increase on the upside here where the wind is blowing toward, relative to the downstream side where the wind is blowing away from? This is even shown with Twitter data that's fun to use, or at least it used to be easier to access and fun to use, and we could see that there's more aggressive and angry tweets happening in areas on days that are having this very high pollution exposure.

Daniel Raimi: It sort of seems like the more we learn about PM2.5, the more troubling it sort of gets in all sorts of domains.

Your paper focuses on the specific context of traffic accidents. Can you tell us a little bit about the data that you and your coauthor use and the methods you use to try to figure out how particulate matter is affecting driving behavior in traffic accidents?

Travis Roach: Sure. My coauthor on this paper is Anne Burton, who is a really excellent health economist down at the University of Texas at Dallas. She has another paper looking at Uber's entrance into various markets and seeing how fatal accidents change. We teamed up, because I had some prior research on particulate matter pollution, and I knew she had looked into this outcome variable.

What we wanted to do is combine this idea to see, Are there more fatal accidents on days that have high pollution? Because the prior research suggests, which it very much does in a lot of contexts and a lot of countries, that you're both more aggressive, and you make more errors in judgment. One of the most costly places this can happen is while you're behind the wheel, something most Americans do every single day, either going into the store or going to work.

It's one of those shocks that affects everyone everywhere all at once. We wanted to look into this and see, Well, if the received literature here says that you are going to be more aggressive, you're going to make more mistakes, does that show up in these fatal outcomes? What we'll eventually find is yes, it very much does; by being exposed to higher pollution levels, we notice a statistically significant increase in the amount of fatal accidents in a county.

Daniel Raimi: That's really interesting. Can you put some more meat on those bones? How big is the effect? Any other details you want to share?

Travis Roach: Sure. We see in our data across all the cities or counties for which we have data on fatal accidents and data on PM2.5 exposure that there's about population-weighted, about a third of an accident a day. You can kind of think about this as a fatal accident happening every three days. What we see is that, by having more pollution exposure, there's a 1 percent increase in fatal accidents. If there's 35,000 accidents per year, then what we're finding is that there's about 350 accidents that could be due to a single unit change in PM2.5. For instance, it wouldn't be surprising to have a day where air quality is in this 9-to-10-micrograms-per-cubic-meter range. That would be considered basically healthy by current US Environmental Protection Agency (EPA) standards, which we'll talk about a little bit more in a second.

But you wouldn't notice a one-unit increase. You wouldn't see the air look different; you wouldn't feel that. But a one-unit increase is what we're talking about by driving up these fatal accidents. Each unit increase is associated with a 1 percent increase in fatal accidents, which is quite large. If we kind of scale that up to a standard deviation change, that's about a 5 percent increase in fatal accidents.

This is a very meaningful outcome, especially if you're behind the wheel. But if we could think about reduction or harm reduction on the other side, when we move to a new particulate matter standard, that could really make a meaningful change in the amount of fatal accidents that happen per year.

Daniel Raimi: As you said, we'll come back to that in a minute. Can you also give us a little more context on the wildfires, which is certainly something that all of us experienced in the summer of 2023, and that folks in Texas and elsewhere are experiencing more recently? What are some of the particulate matter–level increases that communities might experience from a big wildfire, and how might that map on to these results?

Travis Roach: There's two ways we could talk about that. What listeners are probably most used to is what's called an air quality index (AQI). If you have a weather app, this is probably the one that it shows you, "Hey, maybe don't go for a jog today, the air quality is kind of bad." The normal range for AQI is 0–500, 500 being, "That's a bad day. You need to stay inside," and 0–50 being, “The air quality is good.”

Now, under the old 2012 standards, 0–50 being good mapped on to PM2.5 being in the 0-to-12-micrograms-per-cubic-meter range. The majority of our data in this paper is in that range. There are several counties and several days that bump over that, but that's what we should consider good and normal.

Now, as that scales up—so, moderate is from 51–100 on the AQI scale—and when we get into unhealthy for sensitive groups, that would be 100–150. Now, what that means in micrograms per cubic meter, that's in the 35-to-55 range. That would be a specifically bad day. That's the kind of day that's associated with some of these wildfires. Then, when we get into AQIs of 151 and up, that is specifically unhealthy. That is when we have PM2.5 in the 55 up to 150 micrograms per cubic meter or up to 250 micrograms per cubic meter. That’s really acute exposure to high levels of particulate matter.

Daniel Raimi: That's great. As you were answering that question, I was curious to see some of the real-time air quality in different parts of the country. There's this great service called PurpleAir, where you can check real-time data. Where I live in Ann Arbor, today, the AQI is about 88—so, not great. Where you are in Central Oklahoma, it's in the 60s to 70s today, and down in Austin, Texas, where you grew up, it's over 100.

Travis Roach: Not great. I would recommend everyone in Austin, Texas, take the bus today. At least make it someone else's fault if they get in a wreck.

Daniel Raimi: For a little bit more context, can you help us understand how the scale of effects that you find from PM2.5 compares with the other types of things that can increase the risk of traffic fatalities? I'm thinking particularly about drunk driving, but I'm sure there are other causal factors here too, like weather and stuff like that.

Travis Roach: Weather certainly matters. Drunk driving, like you mentioned … In 2021, there were a little over 13,000 drunk driving fatalities, which is a staggering number.

One thing that we think is neat and interesting about this research is that this is kind of the shock. Economists are always trying to figure out, "Okay, what's the effect of this policy on this outcome variable?" Perhaps, we have the entering of Uber and Lyft that might meaningfully change the amount of drunk-driving deaths, or we might increase the drinking age, or we might give curfews to teen drivers, and we maybe would expect that to change outcomes at the margin.

But reducing particulate matter emissions is something that would affect both drunk drivers and non-drunk drivers. If we are making bad decisions due to our exposure to particulate matter, then maybe that decision to have the extra drink doesn't occur, because you're not making bad decisions in general. We view this as this general equilibrium result where everyone is exposed to both higher or lower particulate matter emissions, and this can improve outcomes even for those people who are choosing to drive impaired, or maybe they're teenagers and they're not great at driving in the first place, and it can really meaningfully reduce these traffic fatalities.

Daniel Raimi: That's really interesting. I wonder now if we can talk a little bit more about policy. I'm curious if you can describe to us the current standards that EPA sets for PM2.5 under the Clean Air Act and how those might change. There's been some rethinking of those standards in recent years. Also—how does that compare to the results you find? Are the standards that EPA is enforcing now consistent with ensuring relatively low levels of particulate matter exposure and reducing the types of traffic accidents that you're describing in this paper?

Travis Roach: It's a really important question. It may turn out in hindsight that one of the most impactful things that the Biden administration does is lowering this PM2.5 standard. Right now, as of 2012, the PM2.5 standard to be considered good is about 12 micrograms per cubic meter, which, prior to us having access to such good data, such good experimental methods, and such good computing power, frankly seemed like a reasonable standard. In fact, it was an improvement. It used to be 15, and it got bumped down to 12.

But what a lot of the literature has seen is that we are seeing some of these negative impacts from high particulate matter exposure, just on average over a year, at a lower level than 12 micrograms. I did some prior research that showed that the threshold was at about 9 micrograms per cubic meter, and several others in the crime literature have seen that.

Moving this new ambient standard from an average below 12 micrograms per cubic meter in a year down to 9, in our case, will very meaningfully change traffic fatalities. But then, long-term exposure for people who will grow up in this world in which the standard is now down at 9 micrograms per cubic meter—they'll have lower lifetime exposure. This could reduce their risk of Alzheimer's; this could reduce their risk of dementia. This is the kind of thing that could change both their life in the immediate term and their life in the longer term.

Daniel Raimi: Just to clarify, I think you mentioned 2012 as the year when the standard went down to 12 units. This level of 9 units—is that something the EPA is currently proposing or just kind of a number that's floating out there in the literature?

Travis Roach: They proposed this in early February—February 7, if my memory serves me right. This was after a years-long discussion: the normal EPA process where they put out potential rulemaking, they think about this, they go, "Okay, what should the new standard be?" They announced in early February that they will in fact be reducing the ambient air-quality standard down to 9 micrograms per cubic meter.

Daniel Raimi: That's interesting. Do we have much of a sense of how that changing of the standard is likely to affect sources? Are we talking about shutting down more coal plants, switching more diesel to electric? Or what are some of the sectors that are likely to be most affected?

Travis Roach: I think the coal-powered sector is one of the first things that will be affected and most affected by this. Where we'll see this most likely have bite is in permitting of new coal units.

Suppose someone wanted to introduce a new coal-fired plant in an area—which is not really happening here—they would face more strict permitting standards, especially in areas … It will take a transition period. There are currently several counties that would be outside of attainment status. If you are currently in that 12-micrograms region, because that was the old standard, moving down to 9 is going to take some changes. Especially in those areas that are kind of on the bubble of attainment and nonattainment, it would be very unlikely that the government, when considering whether or not to permit a new power plant … They have to take that into consideration.

Now, to vehicle exhaust specifically, electric vehicles absolutely will help. There's no tailpipe emissions on an electric vehicle. However, where they get the electricity source, of course, matters. Then, add to the fact, though, that electric vehicles tend to be heavier, so their tires wear out a little bit differently. The tire dust can turn into its own little particulate matter. I don't want to make it feel like, "Oh, that will be an easy solution. We'll get there quickly." There's no free lunch, as the economists like to say.

Daniel Raimi: It's also worth thinking about environmental justice implications here, too, because this is an ambient standard. It would account for the cumulative burdens that communities experience in places like the Gulf Coast or other areas where there's lots of industrial facilities. Do you see it as an environmental justice issue, or is the pollution so dispersed that it has kind of broader effects than just on environmental justice communities?

Travis Roach: It’s certainly an environmental justice issue here. The housing around our most polluting areas tends to be cheaper. There's a revealed preference that people do not want to live next to a power plant. Well, if there's housing that is cheaper, that tends to be lower-income folks who are living around that area, so they're specifically burdened by a lot of the environmental pollution we've seen thus far in our country's history.

But one of those hopeful results that has been shown in this literature is that the Clean Air Act and its amendments have really reduced the burden on, specifically, lower-socioeconomic folks who happen to be living in these areas that have high industrial activity.

Daniel Raimi: One more question, and I don't know how far outside of your expertise this one is, but I'm just really curious about it: How much do we know about what's actually happening inside of our bodies that leads to these negative cognitive impacts? How does the particulate matter get into our bodies? Do different kinds of particulate matter affect us in different ways? If it's tire dust versus coal ash versus something else, does that have different effects? What do we know about the physiology here?

Travis Roach: I'll try and do my best to put my biologist hat on here. Part of your question there was asking about the speciation of the particulate matter. It's a tiny, itty-bitty particle, but what is it actually made up of? That's kind of our research frontier. Is it made up of a heavy metal that's a result of the combustion process at a coal plant? Or is it a little piece of rubber from a tire? I think that is something that we're still looking at and learning about. But as far as how it gets into your body and then bioaccumulates within your body—you breathe it in, and these particles are so tiny that they're able to exit the lungs through the blood.

When it's in the blood, it is taking up valuable real estate there, so this leads to oxidative stress in your brain. If your brain is suddenly getting less oxygen than it otherwise would, you're going to make a few more mistakes, you might tend a little bit more toward aggression. That's where we see some of these immediate-term effects happening,because the particulate matter is coming in and messing with your blood-oxygen levels.

The more long-term issues seem to be with this crossing the blood-brain barrier and these tiny particles getting lodged into folds of the brain and various other parts of your body. That's why we're seeing this effect. We see increased Alzheimer's is one of the big results here; cognitive impairment in your older age really starts to go down, in part just because the particulate entered the body, traveled through the bloodstream, and got stuck.

Daniel Raimi: When it's stuck, does it sort of disintegrate? Does it stay there? Do we know much about that?

Travis Roach: If we do know about it, that's something I don't know about.

Daniel Raimi: Interesting. A podcast for another day.

Travis Roach: Yeah.

Daniel Raimi: Well, Travis, this has been really interesting. There are so many interesting studies on PM2.5 that we've talked about today, and your paper definitely adds to the literature on such an important topic.

I'd love to ask you now about what you're reading or what you're watching or what you're listening to, something that's at the top of your literal or your metaphorical reading stack that you think is great and that you'd recommend to our listeners.

Travis Roach: I think I'm going to do a little bit of a left turn from what folks often suggest here. This is, I should say, one of my favorite parts of the podcast. I've gotten so many great book recommendations over the years. What is literally on the top of my stack right now is a book called Prairie Up: An Introduction to Natural Garden Design by Benjamin Vogt. I think my hope here is to maybe inspire people who are thinking about picking a book—maybe pick up this book, but then also get out in your yard and go plant some native plants.

Specifically, this is a passion project of mine. This is what I like to do in my free time, but I do think there's this barrier to entry in, How do you get started with natural gardening? This is the perfect book to help you get there. How do you use native plants? What are the different types you can do, and how to make it also look beautiful? I think (a) go pick up this book, but then (b) maybe go pick up a shovel and carve out a big section of your lawn and turn it into something that's very useful for the wildlife around you.

Daniel Raimi: That's really cool. I'm just looking this up. It looks like a really fascinating book. Actually, my wife and I and our five-year-old are thinking about what we want to plant this year as spring is coming around. It's a timely and an excellent recommendation.

Travis Roach: Awesome. If I can add a couple more recommendations here, just because I’ve done this a few times.

Daniel Raimi: Go ahead.

Travis Roach: You're going to go out, you're going to mark out the area you want to add, whatever that is, you should probably double its size, because this is a job you really only want to do once, digging out any kind of Bermuda that might be there. Then, I think this is probably the kind of thing that a lot of researchers and that type of person that listens to this podcast, the kind of thing that you can scratch that itch—go find some good native plants by your ecoregion. There's a ton of sources online to buy these. I'm thinking Missouri Wildflowers is a good one; Prairie Moon Nursery, Prairie Nursery, Iseli Nursery. You can get these online, or probably, hopefully, locally to you.

Pick yourself five plants and just repeat these in this bed. It doesn't have to be a very complex thing. Pick two or three grasses that are native to your region. Pick an aster that's native to your region. Pick a goldenrod or a sunflower, a Helianthus that's native to your region, and then maybe do something for a special butterfly. Plant something for the swallowtails. Plant something for the monarchs. You're thinking about five or six plants at all, and then just repeat them in your area.

Daniel Raimi: That sounds like a lot of fun. Awesome.

Well, Travis Roach, one more time, from the University of Central Oklahoma, thank you so much for joining us, for sharing your research on PM2.5 vehicle crashes, and your suggestions on gardening tips. It's been a really fun conversation.

Travis Roach: Awesome. Thanks so much.

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 a 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 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.