Search the Site

Episode Transcript

You may not want to do this, but let me ask you to rewind your memory to March, 2020, when Covid-19 was starting to lay us all low. We were hoarding toilet paper; we were wiping down grocery deliveries. Some people, meanwhile, were taking on bigger challenges.

Amy KIRBY: My name is Amy Kirby, and I am the program lead for the National Wastewater Surveillance System at C.D.C. 

To most people, “wastewater surveillance” may not seem particularly germane to a global pandemic. But most people aren’t Amy Kirby.

KIRBY: People will sometimes look at me and say, “Well, you just think wastewater is the answer to everything.” And I’m like, “But it is the answer to a lot of things!”

Today on Freakonomics Radio, three surprising stories about water. The first one is microbial; the second is racial; and the third, geopolitical:

Natalie KOCH: This is not a problem of the Saudis coming and stealing Arizona’s water. This is a problem that Arizona has bad water policy.  

Water, water everywhere — but you have to stop to think.

*      *      *

Amy Kirby is a microbiologist and epidemiologist who grew up in Atlanta. That’s where the C.D.C. — the Centers for Disease Control and Prevention — is headquartered. When it was established, in 1946, C.D.C. stood for the Communicable Disease Center; its main goal was to fight the malaria that haunted southern military bases. Today, the C.D.C. is a federal agency under the Department of Health and Human Services. Amy Kirby always knew she wanted to play for the home team someday.

KIRBY: I have always been interested in medicine and infectious diseases and how these tiny microbes end up causing such dramatic symptoms in people. And I knew that if that’s what I wanted to do, the place to be was C.D.C.

She started out in academia, teaching at Emory University. Eventually, she got her chance.

KIRBY: In 2017, I came to C.D.C. to work in the waterborne disease prevention branch, and to start a program looking at monitoring antimicrobial resistance in water. And then in 2020, when the Covid response started, I was deployed to the Covid response.

Stephen DUBNER: Can you just tell me what it was like to be at the C.D.C. as Covid is starting to blanket this country, and you were being redeployed — describe what going to work was like then. What was actually happening? What was the emotion and mood like?  

KIRBY: Oh, man, I’m glad we’re not on camera because I’m probably going to cry. It was — it was really intense. C.D.C. has some of the most-mission driven, dedicated people I’ve ever met. And so we were working ridiculous hours, trying to do anything we could, sometimes against some very strong headwinds. And then you would turn off your computer and go home and worry about, what is my kid going to do? What — my husband is a police officer. He has to go out and work and be face-to-face with people. You know, what’s his risk? How do I keep my parents safe? There was never a break from it. It’s one of those places where I’m not sure that having more knowledge actually kept you from being less scared. A lot of times that’s true, but in this case it really wasn’t, because we knew what we were up against, and how hard it was going to be, and how scary it was. 

Kirby’s team was hoping to somehow get ahead of Covid and slow down its spread. As wise people often do during a crisis, they looked toward the past — in this case, polio.

KIRBY: So polio, although we think about it as this paralytic disease it’s really a G.I. — a gastrointestinal disease, right? So it infects your gut. It is shed in feces. You get it by eating contaminated food or drinking water that’s contaminated with the virus. So it looks a lot like E. coli in that respect. But only a very small percentage of people, like 1 percent of those that are infected with the virus, will actually develop paralytic polio, which is what we’re most worried about. So when you’re thinking about a signal for surveillance, you’re losing that 99 percent that are infected. They’re contagious, but you’re not going to see them in your surveillance. 

In the 1930s, the Yale epidemiologists James Trask and John Paul proposed that it might be useful to study wastewater — raw sewage — to track down that 99 percent. At the time, there wasn’t a reliable chemical test for the polio virus, so they used a different method: ​​they injected wastewater into monkeys and then waited to see if the monkeys developed polio. Fortunately for monkeys everywhere, the march of science would bring more effective methods of testing wastewater.

KIRBY: And so by looking at wastewater, we gained back that 99 percent. And that has been used for polio for decades. It’s helped us to eliminate polio, and really direct the vaccine where it needed to be.

DUBNER: How would you describe to a layperson the overall benefits of wastewater surveillance?  

KIRBY: The biggest benefit of wastewater surveillance is that it allows us to monitor the health of a community at large without asking the people in that community to do anything that they’re not already doing. And that’s huge. 

DUBNER: It’s huge because why?

KIRBY: Because usually we’re asking them to go get tested, to agree to have that test sent up to the state health department, to give information about where they live and where they work, and there’s often a lot of P.I.I. that goes — Personally Identifiable Information — that goes along with clinical testing. And we don’t have to ask them to do anything. In fact, we don’t want them to. We want them to keep their behavior exactly the same. And we can tell them, now look at this information and say, you know, if Covid’s going up in your community — do you want to take extra steps to protect yourself? Do you want to wear a mask? Do you want to not go to that large event? You know, use it like a weather report, but for the health of your community. 

In the case of Covid, there were a variety of reasons that wastewater surveillance might be useful. Access to Covid testing was still limited. Not everyone who had the virus showed symptoms; they might be contagious and not know it. So: there was potentially a lot of upside to creating a robust Covid-alert system.

KIRBY: We were lucky that we had a study that was already funded to look at antimicrobial resistance in wastewater, by Sean Norman at the University of South Carolina. It was just wrapping up, and I reached out to him and I was like, “Any chance if we could fund you, you could add Covid to this, and keep going?” So he was able to collect data from that wastewater, and we were able to look at that compared to clinical data in that community, and show that if you measure SARS-CoV-2, the virus that causes Covid, in wastewater, you can detect those clinical trends four to six days earlier. That sounds like a short amount of time, but in 2020, that was an enormous amount of time. 

So this one trial seemed like a proof of concept. But Kirby wanted to build out a national system.

KIRBY: Yeah, we had to build it all from scratch. There was nothing — which made it really hard to sell initially because it wasn’t like we could leverage an existing system. There were no systems where we were routinely sampling wastewater for infectious diseases and it was connected to public health. There was none of that. 

And remember: this was a confusing, intense time for everyone, but especially if you happened to be at the intersection of public health and politics — which the C.D.C. is.

KIRBY: Leadership overall had a lot of questions. You know, even with as much money as went into Covid response, we needed to be careful and spend it on the right things. And so there were a lot of questions about the amount of risk associated with this project. Is this something that’s going to pay off? Is it going to be a worthwhile investment? Do we think this data is going to be useful or is it just going to tell us what we already know? And so we really had to prove that it has value, that it’s accurate. And it kind of changed from day to day. Sometimes we would get a good response and they would think, yeah, this is something worth pursuing, let’s do it. And then the next day we’d get a little blowback that we shouldn’t do that. And just combine that with the emotion of the pandemic, and it was a roller coaster for sure.   

Finally, Kirby’s team did get the go-ahead from C.D.C. leadership. They named their project the National Wastewater Surveillance System, or “news.”

KIRBY: All good government programs need to have a pronounceable acronym. So as soon as we hit on “news,” we knew that was the winner. 

DUBNER: Can you describe —  in a way that a layperson can understand — the specific mechanism by which Covid, or the SARS-CoV-2 virus, gets into the wastewater, and then how it’s detected? 

KIRBY: So, we know that we can detect viral RNA for SARS-Cov-2 in stool of people that are infected, whether they’re symptomatic or not. We don’t really understand how it gets there for a respiratory infection. There’s some thought that people are swallowing, you know, like respiratory secretions and it’s coming out in stool. There’s also some evidence that it can actually infect your gut, and so the virus is replicating in your gut, even though it causes no symptoms. We don’t totally understand, is the bottom line. But it makes a really nice signal for us to track who’s infected. And so people will use a toilet that’s connected to a sewer system, and when they flush their toilet, on average in the U.S., it only takes three hours to get from a toilet to a wastewater treatment plant. So when we take that sample at the wastewater treatment plant, it’s very representative of that community that day.  

DUBNER: And then how much time from the sample at the plant until it’s actually analyzed by your team?

KIRBY: Most of our samples are shipped overnight to a laboratory. They don’t actually come to C.D.C. We use a distributed laboratory network, because it’s more efficient. So, they’re overnighted to a laboratory, and then the test itself takes 24 to 36 hours. And then the data is uploaded to C.D.C. for analysis, and it’s available to the health department as soon as we have it. So, from the time the sample is collected until data is available is, on average across our 1,400 sites, about five days. 

Kirby and her colleagues at the C.D.C. found that wastewater surveillance was an incredibly reliable tool to track the spread of Covid over time and place. In the process, they also built a large, live database. And, it turned out, other scientists could look at wastewater data and find other patterns — even stranger patterns.

JOHNSON: So, we started finding these lineages from wastewater that, as far as we know, didn’t exist. There’s a sample that came in from a sewer shed in St. Louis which had this combination that looked absolutely nothing like anything we had seen before.

That is Marc Johnson.

JOHNSON: I am a professor of molecular microbiology and immunology at the University of Missouri School of Medicine. 

DUBNER: Your Twitter handle is @SolidEvidence, correct? 

JOHNSON: That’s right. Uh-hum.  

DUBNER: Explain, please. 

JOHNSON: Well, I started the Twitter account only because I was trying to get people to send me their s**t. I was trying to reach out to groups to get more wastewater from more places. And it kind of snowballed. I had people all over the world sending me their wastewater.

Before Covid, Johnson worked mostly on HIV. But now he was caught up in the SARS-CoV-2 virus, and especially that one sample in the wastewater from St. Louis.

JOHNSON: It was driving me mad. I couldn’t sleep at night. I was like, “What the hell is this?”

Biologists call a sample like this a “cryptic lineage.” As you may remember, Covid variants were originally named after letters of the Greek alphabet — there was the Delta wave, the Omicron wave. If you’re wondering, we are now onto the JN.1 variant. But these “cryptic lineages” that Johnson and his colleagues were identifying — they were strains of Covid that had never before been observed in patient samples. So they started tracking cryptic lineages in wastewater data from other parts of the country, including New York City.

JOHNSON: We kind of had three theories. The one that I thought made the most sense is that it was coming from some kind of domestic animal. So, my likely hypothesis was that it was coming from like a dog shelter. But then we sort of narrowed down what part of the sewer shed it was coming from, and there weren’t any dog shelters. There were no animal shelters, there was nothing that fit. So, our second theory was that it was coming from a wild animal, so the most likely culprit was rats. But then we looked at all of the sewer sheds where we had found these weird lineages, and some of them didn’t have any rats. The final hypothesis — which I never really considered — was that it was coming from individuals. I just didn’t think it was possible that this much signal could come from one person. 

Here’s how Johnson became convinced that one person was indeed capable of generating that strong of a Covid signal. It happened when he got some samples from the Wisconsin Department of Health Services.

JOHNSON: Wisconsin sent me samples from, I think, eight different sewer sheds. And one of their sites had this really hot and very — I don’t know if juicy is the right word, but I mean, it had more mutations than Omicron did. And it was like, “Wow, this is interesting.” And so, we started getting samples from sub-sewer sheds. There’s several lines that fed into the main sewer-treatment plant, and only one of them had the lineage. And so, we’re like, “Oh, let’s keep going.” And so, we checked all the pump stations that contributed to that. And once again, only one pump station had it. And we’re like, “Well, let’s keep going.” And so, probably 50 manholes along the way, and we could follow the signal straight up the line. And then we eventually got out of the city and into sort of a village area. Started narrowing it down to neighborhoods and chasing lots of false leads because we had theories about where it would come from and my colleague would go to dog parks in the area: “Excuse me, sir, may we collect your dog’s feces?” There was a senior citizens’ home. Like, “That must be it!” Nope. Negative, negative, negative. We eventually narrowed it down to a single manhole. And it was clearly — you know, one more upstream, there’s nothing. But it starts here and then everything downstream, it is present. 

DUBNER: And this is like, a manhole that’s connected to one building or something? 

JOHNSON: It was. There was only one input to this manhole. And that input was from not a nursing home, not a shelter. It was a company. Small company, has about 30 employees. At this point, the health department was really involved. I was just doing the sequencing to tell them what was positive and negative. Then they went and they contacted the company and said, “Okay, we think someone in your company has a very long Covid case that they might not know about, and we would like to offer them free testing.”

DUBNER: Were they a little flipped out, this company?  

JOHNSON: They were actually remarkably calm. This was a ways into the pandemic. I mean, people were not as scared as they had been in the beginning. They had two sets of toilets in the building. One was only used by their employees and the other one was used by the public. And I think they were sort of hoping it would come from the one by the public. And so, we tested both lines and we got the line from their employees and it was only there. This was the game-changer because I could not believe how much Covid there was in that sample. It was 1,000 times higher than anything I had ever seen. There was still so much Covid in that wastewater, we could detect it with a rapid antigen test, one of those little, you know, things where you put two drops of your snot and it gives you a red line. And those things are not very sensitive. Once we narrowed it down, we did communicate with the employees — they, I wasn’t involved with the communications — and about two-thirds of them got free testing by nasal swabs, and they were all negative. So, it could be the person who had it was not tested or it could be that the virus just wasn’t in their nasal cavity anymore. 

DUBNER: If you could have identified and interviewed that person, what would you have asked them?

JOHNSON: Number one, I’d like to know, what are your symptoms? Do you have G.I. issues, do you have other issues? And the other thing is, do you have underlying conditions? Up until yesterday, we have now found 43 of these cryptic lineages. But I’m sure what we’re seeing is the tip of the iceberg. What really drives me now is that — is there a connection between these persistent infections and long Covid? We’ve never actually tracked one of these to a single person. So, we can’t say with certainty that this is a threat. It’s sort of a Catch-22 that you need more information to figure out if it’s causing problems. But if there’s no evidence that it’s causing problems, then you can’t justify a bunch of public health to figure out who these people are. 

And even if you can figure out “who these people are,” there’s a new set of questions to think about. For instance, how would you feel about a health department official knocking at your door with exciting news about your stool? Or maybe it’s the police who knock, and maybe it’s not about your stool, but something else in your wastewater.

JOHNSON: Let’s say there’s a meth lab that dumps a whole lot of stuff down the drain, yeah, you could use it for tracking it down. I mean, there’s a reckoning that needs to occur. There needs to be conversations about. What is the public going to be okay with us doing? I mean, already there’s states that have tried to put in regulations banning wastewater testing for this.

DUBNER: Oh, really?

JOHNSON: Yeah, I think North Dakota had a bill that had been submitted many times. And I think that’s a lot of resistance that health departments have, is that this is a really valuable tool for them. They don’t want to lose it. They don’t want there to be a backlash about what we’re doing it so bad that they can’t do the useful parts anymore. 

KIRBY: I mean, there is a lot of human DNA in wastewater.

That, again, is Amy Kirby, from the C.D.C.

KIRBY: Feces contains a ton — more human DNA than almost anything else. But we very explicitly do not measure that. Because the goal here is not to identify people. It’s a public-health tool. Our methods right now are very specific to SARS-CoV-2. We have some sequencing that we do to look at variants. It’s also specific to the virus, but you can get off-target measures that will detect little pieces of human DNA. And the first thing we do before anyone ever looks at that data is we put it through what’s called a Human Read Scrubber, and it pulls out all of the human reads, so that we don’t ever see them. And so we’re taking a lot of care to not see that. 

DUBNER: Right. But theoretically, if I got hold of your data or if I hacked into a wastewater surveillance system, I could measure that. What would you see as being the dangers or privacy invasions of that kind of data being analyzed?  

KIRBY: You know, with enough sequencing, you could identify people right, out of wastewater. 

DUBNER: Individuals, you mean.

KIRBY: Yes. It would be hard. And you are digging for a needle in a haystack. But is the technology possible? Yes, it is. And so what I would worry about are things like identifying someone that has cancer, and is that going to impact their job, their insurance. I worry about things like law enforcement wanting to use this to find fugitives. I mean, I can see why that would be an attractive thing to do, but it’s not what we want to do for public health. And so that’s why we’re putting up very hard guardrails at C.D.C., that we do not keep any of the human data. Because we want to make sure to have the community’s trust in this system. So if there’s enough pushback from the community, they’ll stop, the utility will stop sending samples. It’s the communities that are rejecting all things Covid, you know, we will often see those communities also pull out of wastewater surveillance.

DUBNER: What would you say would be, from your perspective, the dream scenario of your project — let’s say, five years from now? What data would you have it gathering and what purpose would you see that data serving?  

KIRBY: Oh, this is fun. So, a couple of things. I would want to have every wastewater treatment plant in the U.S. participating. That’s about 80 percent of houses in the U.S., are on sewer. 

DUBNER: And where are you now?  

KIRBY: We’re at 40 percent now, so we’re halfway there in three years, which is pretty good. What amazes me is not so much about Covid, but as we’re thinking about expanding the system to other targets, now that we’ve built this huge platform, is how many health issues can be monitored through wastewater.

DUBNER: Like what?

KIRBY: We’ve gotten questions like, “Can you detect West Nile virus in wastewater?” My gut instinct was, no, that’s not going to work. And then we started looking and sure enough, you probably can. Tuberculosis is another one that I got asked about. And I was like, no way. Turns out you can, and so we’re looking at that. Influenza, RSV — so there’s all of these things that we can look for. We can look at cortisol: how stressed is this community? We can look at caffeine: how much are they relying on that? We can look at illicit drug use. We’re also thinking about how NWSS and wastewater testing can be part of nationwide preparedness, right? So when the next emergency comes, is the system ready to respond? That’s really the goal. So we’re going to be doing surveillance every day for the things that we know are always out there, but also being ready to respond as quickly as we can, should there be another — heaven forbid — pandemic or some other type of massively disruptive response.

Coming up: here’s a puzzle that might be really hard to solve:

Brian BEACH: The largest decline in the Black-white mortality gap occurred during the height of Jim Crow.

That might be hard — if you weren’t a water researcher.

*      *      *

Brian Beach is an economic historian at Vanderbilt University, in Nashville.

BEACH: Water is definitely one of my specialties. 

With a name like Beach, you might think he’d study the cost of ocean desalination, or the benefits of harnessing wave power to generate electricity. But no, Beach works inland. He’s focused on what comes out of your kitchen tap, and what goes into your toilet. He is enthusiastic about his work.

BEACH: I got into water primarily through my advisor, Werner Troesken, who was at the University of Pittsburgh. 

Troesken died in 2018. He and Beach had co-authored several papers about municipal water systems.

BEACH: Ideally, water would be something that we don’t think about. Like you can turn on the turn on the tap and enjoy it and it’s safe and it’s available. But that’s, of course, not always the case. 

DUBNER: And that’s the input. What about the output? Do we think even less about where our waste goes?  

BEACH: Ideally, yes. We would think even less about where our waste goes. However Werner, I think, had a real knack for getting people excited about sewage, and all the other disgusting things associated with it.  

Many of the municipal water and sewage systems across the U.S. were built in the late 19th and early 20th centuries. And before that?

BEACH: You would have a collection of wells dispersed across the city, and you might go and fill up a pail and bring the water into your household and then store it. And this is an issue, of course, because with storage, you might have flies or something like that, that could further contaminate the water supply. And then in terms of waste, a lot of the waste was just disposed of basically in your backyard in a cesspool. And if it’s not watertight, then any of your sewage could leach into the soil, and then ultimately find its way into the well. And so this sets the stage for the fecal-oral transmission of typhoid fever.

DUBNER: From where we sit now in the 21st century, give us a sense of how widespread and dangerous waterborne disease used to be — maybe sticking with the typhoid fever example. 

BEACH: So, if we think about Covid, and the Covid death rate in 2020 or 2021, you’re talking about a death rate on the order of about one death per 1,000 persons. And this is obviously a big deal, and we changed everything that we did for a few years and then maybe forever. And if you think about just typhoid fever, in a particularly bad year, you might have a death rate that’s like one-and-a-half deaths per thousand persons. In a good year, it might have been like half of Covid. 

DUBNER: So on average, in the neighborhood of Covid at its worst.

BEACH: Yeah. And it was just year after year after year. It was like the 15th-leading cause of death in 1880, and then by about 1930, we had largely eliminated it. It’s probably some of the most compelling graphs that you can make, is looking at the year-to-year typhoid fever rate in, say, Pittsburgh. You know, Pittsburgh had some of the highest typhoid fever rates because it drew all of its water from rivers, and they were heavily polluted. And then after they built the filtration plant, you basically eliminated typhoid fever in Pittsburgh. 

Building a city-wide water and sewage system was expensive.

BEACH: For most of these cities, these are going to be the largest investments that any of the cities would have made. New York City had a pretty extensive system because they had to bring water in from an outside source, and so they had to transport water, I think it was like 40 miles. And in per capita terms, it was much more expensive than building the Panama Canal. 

At the same time that all this water infrastructure was being built in U.S. cities, the American South was still enforcing racial segregation via Jim Crow laws. So as cities in the south were building new water systems, how did segregation shape their decisions? That was a question asked by Werner Troesken, Brian Beach’s advisor, in a 2004 book called Water, Race, and Disease. I remember reading that book when it came out, and being so surprised by its central claim; that’s the reason we’re talking to Brian Beach today.

BEACH: Water, Race, and Disease is a fantastic book. And in that book, Werner documents a paradox of sorts. The paradox goes something like this, which is that if you were to look at a measure of racial inequality — and you could use health or life expectancy, mortality, anything like that — if you were to think about where the largest gains in health were, the closing of that racial health gap, it wouldn’t actually be after the civil rights movement. The largest decline in the Black-white mortality gap, or largest improvement in the Black-white mortality gap, occurred during the height of Jim Crow. So that’s puzzling because this is a time when Black households, Black individuals in general, were excluded from effectively every other public service. And so what Werner goes on to describe is a situation where between 1880 and 1920 and maybe a little bit beyond that, that’s a period where cities are dramatically expanding their water infrastructure. And the story he tells is that if you’re building a water network or a sewer network and you’re digging up the streets, it’s kind of silly to connect only the white households and not the Black households. And because of this, and because water was so important for health, what you actually have is a situation where the — I guess the economic incentives were such that despite the racism that was prevalent at the time, you actually had relatively equal access to these supplies.  

DUBNER: When you say it was “silly,” my reading of it was, it wasn’t so much that it was like, “Well, we’re here, we might as well connect, you know, the neighborhood to the left as the neighborhood to the right.” But part of it was about white self-preservation — like, we are making a sewage system here to protect our white populations. But also, we don’t want the Black population’s sewage to contaminate our water as well. Am I misreading that?  

BEACH: No, you’re not misreading that. That’s exactly correct. This is a point that Werner makes quite strongly, is that if I’m going to be drinking water that’s contaminated with sewage, that’s bad enough. It’s even worse if it was contaminated with sewage from the Black neighborhood. I would say that the self-preservation part of the story is a little bit more nuanced, which is something that I think we’ve come to learn from some of my follow-up work in the recent years. 

DUBNER: You have written that “water and waste systems are among the most crucial pieces of urban infrastructure. Many people remain perplexed as to why a country as rich as the United States would allow its infrastructure to fall into disrepair … The situation,” you write, “is less puzzling, however, once one recognizes that the history of urban water and waste management in the United States has always been more reactive than proactive.” What do you mean by that exactly — especially because it sounds like a lot of this early water-system building was fairly proactive. Or maybe I’m wrong. Maybe it was reactive against, you know, disease in the white communities. 

BEACH: Yeah, Philadelphia was the first modern waterworks in the United States. 

DUBNER: Which was when?

BEACH: It was about 1800. And the motivation for building the waterworks in Philadelphia was actually a yellow fever epidemic, which is not caused by water — it’s mosquitoes. So there was this misinformation, but this is a time when you have, like, the miasma theory of disease, the spontaneous emergence of disease, like we have an incomplete understanding of it. And so people suspected the water because it didn’t taste very good, it smelt poor, things of that nature. About 1880 is when we start to see the widespread diffusion of the germ theory of disease. Cities are mostly investing in their water infrastructure as soon as they start to outgrow it. And so if you trace it and you think, like, who’s building a waterworks prior to 1860, it’s the largest cities in the United States. And it’s simply because they’re outgrowing their supplies.  

DUBNER: What would a generally more proactive drinking water and waste management system look like? What would you like to see ideally, let’s say now in 21st-century U.S.? 

BEACH: I think we would certainly want to pay attention to the water quality. One is going to be the materials that we’re using, so the environmental conditions. I think we’re right to remove lead simply because there’s more and more evidence that there’s no safe level of lead exposure, especially for young children. And I think there’s room for improvement particularly in the reporting of water quality, and the transparency. I think we’re actually moving in that direction and I’m pretty optimistic about it. Where I think we could be more proactive is actually addressing poor quantity of water. So if a water main bursts, then you’re without water. There are some cities in the United States, or more rural areas, that are without reliable, constant access to water. And so I think in that situation, what we would want is to basically figure out who’s being left out and make those investments to try and bring more equality of the access.  

In the past decade, residents of both Flint, Michigan, and Jackson, Mississippi, have been unable to get safe drinking water for extended periods. Flint’s water system has been overhauled. In Jackson, there are still ongoing issues. Both cities have majority-Black populations. There may be a wider crisis in the future. The Environmental Protection Agency estimates that U.S. water systems will need to spend more than half a trillion dollars over the next two decades on water infrastructure — and that doesn’t even include sewage.

BEACH: A lot of our water infrastructure is at the end of its usable life. Yes, we still have the filtration plants, and so we can purify the water. We can alter the chemistry to reduce the incidence of lead exposure. But we are now back in a situation where we basically need to dig up these roads and replace the infrastructure. Someone was just telling me that there was a wooden main in New York that was just replaced, like, this year.  

Coming up: have you heard the one about the Saudi Arabian cows that eat hay from Arizona?

*      *      *

As you may have heard, some parts of the U.S., especially the Southwest, are having big problems getting enough water.

CBS NEWS: A water crisis is unfolding in the nation’s Southwest, as the mighty Colorado River begins to dry up as a result of overconsumption and climate change.

12 NEWS: Another Arizona community is now face-to-face with a critical water shortage.

GOOD MORNING AMERICA: The taps are dry in Rio Verde foothills after the nearby city of Scottsdale cut off the small town’s water to deal with its own drought. 

The Colorado River provides water and power across the Southwest, and the reason it’s drying up can be traced back 100 years. That’s when seven states and Mexico adopted a legal framework, known as The Law of the River, to allocate its water. If you’ve ever seen The Producers — well, it wasn’t quite that bad, but the principle is the same: oversubscription. In the case of the Colorado River, more water was promised to more places than could sustainably be delivered. In 2023, the federal government cut a deal with some states, including Arizona, that would pay them to use less water from the Colorado River. Arizona, one of the hottest and driest and fastest-growing states in the U.S., also announced that it will limit the construction of new homes around Phoenix because of low groundwater levels. So it may be surprising to learn that one major consumer of Arizona groundwater is a huge dairy company from Saudi Arabia called Almarai. They are headquartered in Riyadh, and are associated with the royal family. In 2014, they began buying and leasing land near Butler Valley, northwest of Phoenix, to grow alfalfa to ship back to Saudi Arabia to feed their cows to make yogurt and cheese. And why, you might ask, does a Saudi dairy company find it worthwhile to grow its cow feed all the way over in Arizona? For that, we need to bring in an expert, who also happens to have grown up in Arizona. Natalie Koch is a geography professor at Syracuse University, and an author:

KOCH: The book is called Arid Empire: The Entangled Fates of Arizona and Arabia

Koch’s academic focus is an uncommon one.

KOCH: I look at authoritarian regimes that have a lot of oil and gas money. 

And, more specifically:

KOCH: I’m interested in how it is that certain authoritarian regimes are understood to be popular and legitimate. We have this really simplistic understanding, I think, in democratic places that they must just hate their government. That’s not true at all. And so I’ve always tried to understand how it is that that works. How do some people in these systems really like them? How do they get pulled into the system? And who is helping with that process? When you look at the places where I work, primarily in the Arabian Peninsula but before that also in Central Asia, you have a lot of Western experts — consulting companies, engineering companies, architecture firms. So I’m sort of interested in the politics of how those relationships continue to sustain this world that we live in, which we just imagine it’s divided between authoritarian and democratic countries. 

Okay, so getting back to Almarai, the Saudi dairy company — what makes Arizona such a special place to grow their cow feed?

KOCH: What makes Arizona special is that when you have this desert land, you can grow alfalfa at an incredibly fast rate because of the amount of sunshine. So, as long as you have plentiful water, you can just grow. So if you took a comparable farm in Minnesota, you would get maybe two or three cuts per year. But in Arizona, you can get ten or more cuts per year. And that’s the attractiveness of the Arizona farms.

Ah, yes, “as long as you have plentiful water.” Okay, let’s go back in time to see the genesis of the relationship between Arizona and Arabia. There once was a man named Karl Twitchell.

KOCH: Karl Twitchell is quite a character, let’s say.  

He was an American mining engineer.

KOCH: He worked in the copper mines for a couple of years, and then he got an opportunity to work in what we now know as Yemen. Early on in his time in the Arabian Peninsula, he got in touch with the handlers of King Ibn Saud — so, the first Saudi king that gives the country its name. 

The key event to consider here is the discovery of oil in Saudi Arabia. At the time, the U.S. was easily the biggest oil producer in the world, and American oil companies sent engineers around the world to look for more. In the late 1930s, Americans found vast amounts of oil in Saudi Arabia. If you are the kind of person who likes to pinpoint one historical event that wound up driving thousands, millions, trillions of subsequent events — the discovery of oil in the Middle East is a good candidate. It made a few nations unimaginably wealthy, and reshuffled the deck on geopolitics. Karl Twitchell, meanwhile …

KOCH: So he would talk to Ibn Saud about anything related to his experience in Arizona, in the U.S. Southwest, and how similar they were. 

How similar they were desert-wise, for instance.

KOCH: He liked to promise things. He promised the king that he would bring a team of U.S. experts to help him develop the entire country’s agricultural enterprise. 

Because, even though you’re located in a desert, if you have enough oil money, you might as well find a way to grow your own food rather than import it all. This began to happen on a farm about 50 miles south of Riyadh.

KOCH: So the Al-Kharj farm was started with the idea that it would be for the King and the royal family to profit from. 

DUBNER: “To profit from,” meaning as a business or just for resources for themselves?  

KOCH: Resources for themselves. That was the initial impulse. But of course, when once they started trying to get U.S. government money and other actors involved in it, they had to shift the story, and that shifting was more along the lines of, “Well, if we if we can build this wonderful model here, everybody else in Saudi Arabia can benefit from it.” This, of course, is a very typical strategy of any kind of elitist project, is just call it a model, which nobody else has the resources to do. But, you know.

It was Karl Twitchell who had helped get the U.S. government involved in this model farm project. In 1942, the Roosevelt administration sent an agricultural mission to Saudi Arabia. In 1947, the king’s son, Saud Al Saud, took a tour of U.S. agricultural regions.

KOCH: The crown prince really liked the dairies that he visited in Arizona. And so once he got back to Al-Kharj in Saudi Arabia, and he became king shortly after that, he then took charge of the Al-Kharj farm. At this point, it was being handled by Aramco, the Saudi oil company. Well, now it’s Saudi, then it was American. He put the managers in charge of creating a Grade-A dairy like he had seen in Arizona. It took a little bit of time to scale up, but by the 1970s, it really started to expand dramatically — because also in the 1970s, the Saudi government started to invest a huge amount of money into commercial agriculture, and giving farmers subsidies for setting up commercial farms. So Almarai, the dairy company that has farms in Arizona now, they were one of the early beneficiaries of these 1970s-era policies of subsidizing the expansion of the domestic agricultural industry.  

DUBNER: Where is the output of these farms going to? Is it mostly consumed domestically or is some of it for export? 

KOCH: A lot of it is domestic consumption, but it’s also exported all around the Middle East. I first became aware of the issue because Qatar, the country next door to Saudi Arabia, had been put under a blockade in 2017. And one of the first things that the Qataris did when they were put under this blockade was run to the grocery store to get their milk because their milk market was so dominated by the Saudis. 

DUBNER: So as far as I know, dairy farming requires quite a lot of water. Where is all the water coming from in Saudi Arabia, from the 1940s up until the — whenever 1980s, ‘90s, 2000s? 

KOCH: Gulf cities all across the Arabian Peninsula, they cannot exist without desalination. So they almost all run on desalination for drinking water and other basic things. That’s not very economical when it comes to agriculture. So the reason that they wanted to put this big farm initially, the Al-Kharj farm, in that region is that they had relatively easy access to groundwater. But the groundwater reserves in Saudi Arabia have diminished dramatically. So the Saudi government understood very early that they just could not sustain domestic, commercial-scale agriculture. And so they ended, or they prohibited, the growing of commercial-scale green forage, things like wheat and alfalfa that they had been using to feed the cows, that had stopped. The dairy companies themselves, they’ve just had to source their food somewhere else. 

This brings us back to Almarai, the big Saudi dairy producer. By 2014, they had expanded, entering new markets and buying up other companies — and through one of those subsidiary companies, they licensed more than 13,000 acres of farmland in Arizona to grow alfalfa. And that license came with as much groundwater as they could pump. That isn’t the case for all agricultural land in Arizona; this goes back to a groundwater law from 1980.

KOCH: And this essentially divides the state of Arizona into two. There’s parts of the state, counties that are outside of Active Groundwater Management. And then there are the areas that are in the Active Management Zones, the city centers of Arizona. And this was seen as a compromise amongst the farmers who, back when they were developing and adopting the 1980 groundwater law, who really opposed large regulation, that they said, you know, it’s just a city issue. And so the rest of the state that do not have these active management areas, they have to get a permit to drill a well, and then once you have that permit, there’s no monitoring of how much water you are pumping from that well.  

DUBNER: So I realize you’re not an economist, but how do you feel about the price of water in a place like Arizona being essentially zero?  

KOCH: Deeply troubled. I’m from Arizona, I’m from Tucson. And this has always felt incredibly important to me as a child, and understanding that water is scarce and it is painful to see it being wasted in a lot of ways. To know then that big agricultural things like these alfalfa farms continue is really troubling. But it’s only possible because there’s no real cost on it. The challenge that I heard early on when I first started doing this research quite a few years ago from people in Arizona was to say, “Well, if our farmers don’t have the money to drill deeper wells and the Saudis do, then that’s fine. If they can drill deeper wells and we can keep our jobs, that’s great.” However, that attitude has started to shift now because people have started to see — well, if they’ve drilled all these deeper wells, they’re dropping the groundwater level. And so everybody else in the area, their wells no longer reach the water table. And so this creates a huge economic problem for everybody else. 

And this has created political problems too. How could it be that the same state that was restricting new home-building in one place because of lack of water was letting an overseas firm use as much water as they wanted to grow food for their cows?

KOCH: Yeah. So in the summer of ‘22, the Arizona Republic had just issued this investigative report about these obviously problematic lease arrangements with the Saudis. And then you had also the looming Colorado River water cuts. You also then had the election of 2022. And during that campaign season, you had the current new governor, Katie Hobbs, the Democrat, and the current new attorney general, also a Democrat, Kris Mayes. Both of them really attached their campaign to this Saudi farm issue because they understood that people were upset about it, and they didn’t want to imagine that the state political leaders were just giving away Arizona’s water to Saudis. And both of them, since taking office in January of ‘23, they’ve both looked pretty actively for any strategies that they could find to end the Saudi presence in Arizona’s agricultural sphere. This is not a problem of the Saudis coming and stealing Arizona’s water. This is a problem that Arizona has bad water policy.  

DUBNER: Do you think if you and I are speaking five years from now, that Saudi Arabia is still growing alfalfa in Arizona? 

KOCH: I would guess not, at this rate. So, in October ‘23, Katie Hobbs announced that they were canceling one of the leases in the Butler Valley area and that the other two in the Butler Valley area would be terminated in early 2024. So that is something that Almarai has promised to try to appeal and to push back against. 

DUBNER: Here’s something that you wrote recently in The New York Times: “Arizona is not the victim of evil outsiders. It’s the victim of its own hubris and political failings that allows such a system to exist. Blaming the Saudis may be a good political play, but the problems won’t go away until state lawmakers properly reform Arizona’s groundwater laws.” What kind of reforms would you like to see? 

KOCH: They first of all, need to reform this rule that that basically allows for the areas outside of these active management areas to pump groundwater unregulated. So, getting some kind of regulation mandated across the entire state, I think is the first thing that needs to happen. The other would simply be to reimagine how it is that water is paid for by these users. Because if it’s just an issue of counting how much water people are extracting, but they can still extract that water basically for free, then that also doesn’t correct the challenge. So people in Arizona, all users in Arizona, need to be paying a reasonable amount for it, so that you don’t have this just hugely wasteful application of water.

DUBNER: Name some other countries that do something similar to the Saudis with alfalfa in Arizona, if it happens. 

KOCH: Oh, well, missing from a lot of this coverage about the Saudi farms is the fact that the Emiratis have also been growing alfalfa in a farm just next door to the Saudi one. So the Emirati dairy company and agricultural commodity company has been doing the same.  

DUBNER: Well, every Arizonan can feel proud that they are helping develop the Middle East yogurt market, at the very least.  

KOCH: Yeah, apparently. One of the things that I actually was really interested to see was that the Saudi-backed LIV tour, the golf tournament, it came to Tucson earlier this year when I was there. And I had to go because I thought, Well, if people in Arizona are so outraged that the Saudis are coming to use Arizona water to feed the cows, I wonder if there will be any outrage that the Saudis are coming to use our golf courses, which are also incredibly water-intensive and wasteful. Of course, nobody was there. Nobody had anything to say about it. 

That was Natalie Koch, the author of Arid Empire: The Entangled Fates of Arizona and Arabia. Before that, we heard from Brian Beach, of Vanderbilt; Marc Johnson from the University of Missouri; and Amy Kirby, from Centers for Disease Control and Prevention. My thanks to all of them — and to you, for listening. By the way, we made an episode of Freakonomics Radio a while ago, when the Saudi-backed LIV golf tour that Natalie Koch mentioned was just getting started. A lot has happened since then, so we’re going to update that episode for you and play it for you as a bonus. So, keep your ears open for that.

*      *      *

Freakonomics Radio is produced by Stitcher and Renbud Radio. This episode was produced by Alina Kulman, with research help from Daniel Moritz-Rabson. Our staff also includes Eleanor Osborne, Elsa Hernandez, Gabriel Roth, Greg Rippin, Jasmin Klinger, Jeremy Johnston, Julie Kanfer, Lyric Bowditch, Morgan Levey, Neal Carruth, Rebecca Lee Douglas, Ryan Kelley, Sarah Lilley, and Zack Lapinski. Our theme song is “Mr. Fortune,” by the Hitchhikers; the rest of the music this week was composed by Luis Guerra. You can follow Freakonomics Radio on Apple PodcastsSpotifyStitcher, or wherever you get your podcasts.

Read full Transcript


  • Brian Beach, professor of economics at Vanderbilt University.
  • Marc Johnson, professor of molecular microbiology and immunology at the University of Missouri School of Medicine.
  • Amy Kirby, program lead for the National Wastewater Surveillance System at the Centers for Disease Control and Prevention.
  • Natalie Koch, professor of geography at Syracuse University.



Episode Video