Emily Oster

Emily Oster:

If a bunch of people yelling at me online meant that some poor kids got to go to in-person schooling and learn some things, that's such an amazing trade. I'd take that a million times over.

Ted Roosevelt V:

Welcome to Good Citizen, a podcast from the Theodore Roosevelt Presidential Library. I'm Ted Roosevelt. Today my guest is Emily Oster. She's a bestselling author and Brown University economist. Like many of us, she found herself overwhelmed by the conflicting medical information during the COVID-19 pandemic, but unlike most of us, she had the tools to cut through the noise. As an economist, she applied data analysis to answer the questions that so many of us were asking. She rose to national prominence during the pandemic for challenging public health guidance on school closures, offering a clear-eyed look at the data when it mattered most. In this episode, we talk about the importance of building a more data literate society and why she actually enjoys being recognized in her local coffee shop. Plus: how does an economist become a public health authority? You're about to find out.

Emily Oster:

I was with some economists the other night and one of them was like, I just think we have a better understanding of statistical methods than any of the other fields, which I think I wouldn't have said it like that. A lot of what we do in public health, everything about nutrition, a huge share of what's about child development—they don't use very sophisticated statistical methods and they are really subject to the concern that there's differences across groups and other things. I think economics tends to be much more careful about what we learn from those kinds of studies and I think I'm bringing that particular lens onto this. It means I'll pick out one or two studies and say like, Hey, these actually did a really good job. I realize there's 65 studies, but 63 of them are garbage and just averaging a lot of hot garbage doesn't improve things. So really you should just look at the two studies that are good and I think that perspective is somewhat different.

Ted Roosevelt V:

What drives you to seeing a space like bad medical information for pregnant women COVID-19? Why do you get drawn into those spaces?

Emily Oster:

I think there's two things. So one, is I really like learning new things. During pregnancy, some of the motivation was just like, I want to know these answers. COVID wasn't quite like that in the sense that it wasn't super fun and positive, but there was an aspect of, I'm really going to learn everything I can about how this is working and I want to know how much COVID is spreading in schools. And if I am not going to be able to learn that from data other people have, I guess the only way for me to learn that is to collect my own data. And so I guess I will have to do that because I am motivated to learn. That's one big piece of it and I think the second is that I want to have impact. I want to have impact on the world and I try to do things that are the best use of my talents to have impact.

Ted Roosevelt V:

Yeah, I think that's key and something we touch on a lot on this podcast, which is to encourage people not just to follow their curiosity but to turn what they discover into positive action. So you have your PhD in economics and you use those skills to become a bestselling author of a book about pregnancy. That's not the most common path. Can you describe how that specifically happened?

Emily Oster:

There's two answers. So my training is in economics broadly, but my areas of study are health economics and statistical methods. So even before I was pregnant, I was a person who studied health and how people make health decisions and thought about data and what we can learn from data and causality and so on. And then I got pregnant and I found that the evidence around pregnancy was not being subject to the same kind of scrutiny that I was subjecting the evidence that I saw in my academic life to.

Ted Roosevelt V:

Right.

Emily Oster:

And many of the problems that I was sort thinking about academically, like "is this relationship causal or just correlation?", that those issues were being often quite aggressively ignored in the pregnancy literature. And so I started doing a lot of work basically in the service of my own pregnancy, trying to think through decisions about prenatal testing and other things like that. Taking the expertise you have in the world and taking it to some aspect of your personal life that is very important to you, I think that's actually a very familiar experience for a lot of people. The part where you then turn into a book, I think that's a bit weirder and I really like the sort of teaching aspect of writing where it's trying to figure out what's the best way to explain this complicated concept to somebody. That's the sort of first moments of how I made that connection.

Ted Roosevelt V:

There's something about that moment of pregnancy where it's a moment of great vulnerability and where data can be highly reassuring and yet I think your experience, and certainly the experience when my wife got pregnant, is that data wasn't always necessarily informative in the way that you want it to be informative. It was sort of misleading and sort of maybe even scare tactic-y to some degree.

Emily Oster:

There was a piece of pregnancy that is sort of all consuming in a way that most of us have not experienced before. And then on top of that, it's for many people their first sustained experience with the medical system. You never really had to make a lot of significant health decisions, and then you arrive at this pregnancy moment where there's tons of decisions to make and there's a lot of information and there's a lot of things that are scary and where what you're being told is "just don't do this, don't do this, don't do this, don't do this. If you do any of these things, you've ruined everything forever." And that freaks people out and also can be very frustrating because it can feel like, "what should I eat? Everything—you've listed everything." And also people will say, "I have terrible morning sickness. The only thing that sounds good to me is hot dogs. I'm losing weight and all I can eat is hot dogs, but I've been told I have to eat something, but also I'm not supposed to eat hot dogs." We almost have made it impossible and we struggle in the information conveyance to help people prioritize and explain, here's the first best and here's the second best, and which of these pieces of evidence are really well supported by data and which are just not very compelling.

Ted Roosevelt V:

How much of the data that's sort of problematic in this way is via the sort of medical establishment versus via all other social media and other sources of information?

Emily Oster:

Both. I think it's probably depends on what you're consuming. I mean I think that most of the stuff that comes out of what we think of as the medical establishment, kind of official guidelines—the issue is not that it's completely made up, but that it is based on sort of correlation, not causation or some kind of vague suggestive evidence. And I think there's a precautionary principle that dominates and makes that system take even very vague, probably flawed evidence more seriously than they should.

Ted Roosevelt V:

You just reminded me of when my wife was pregnant with our second and third, we had twins. The second go around the twins were in separate embryonic sacs. And so when we learned that we were having twins, the OB/GYN—which was sort of this dramatic moment in its own, right>— the OB/GYN said, oh, by the way, they're fraternal because they're in separate embryonic sacs and we didn't think to question— I mean, why would you question that? And when the twins were born, you can see one's bigger than the other. And then we brought them home and when they were feeding their weight normalized and I, sort of being the big dumb dad, was like, I don't know who is who anymore. And at a certain point my wife was like, I'm not a hundred percent sure who is who anymore. And we sent away for a DNA test. It turned out that they were identical. It was a piece of medical information that we had been told as a truth that turned out to be actually a statistical probability, not a definitive truth.

Emily Oster:

I think that happens a lot and I think it often happens for actually very understandable and good reasons, which is that people would like to be told certainty because that is something that people ask for. And the experience of trying to communicate, "okay, well they're in separate embryonic sacs, which means they probably are fraternal, although in some circumstances, depending on when the egg split, they could be identical and here are the probabilities—" That's a much longer conversation and I think that there's a desire to simplify, which doesn't always serve... it doesn't always serve outcomes well, but I think it's pretty understandable and is both sort of supply and demand driven, as economists would say.

Ted Roosevelt V:

When you were pregnant, you started to question a lot of the information out there, realizing that there were many rules and fears that you were being told about that were maybe overstated. And you went back to the data and that helped you better understand the risks involved. Do I have that right so far?

Emily Oster:

Yep, absolutely.

Ted Roosevelt V:

Okay, so I'm curious how you see the role of data more broadly. Do you think it's generally helpful for people to have all the data you had, which often comes with a ton of nuance and complexity, or is it clear messaging? Is that sometimes more important than all the nuance?

Emily Oster:

On the one hand, if you had a setting in which you were sure about the data, it was a hundred percent incredibly confident, definitely correct, and you knew that everyone's preferences were the same, then it would actually be fine to give a sort of "do this, don't do that" messaging. "Don't open the door of an airplane and jump out." Okay? Because we know: mostly people don't want to die and we have pretty good data that, like that's not a good— you know, it's like, "okay, fair enough." There are some things like that and there are some less extreme forms—but where I think we often will sort of push out more certainty than we have is all of these other places where you'd like to acknowledge what the data really says to let people make their own choices because they will make choices that are better for them and because you will help them make better choices. Let me give you a very concrete example of this. It's about co-sleeping. So co-sleeping with an infant is something that there is very strong messaging against in the U.S. So, people are told if you co-sleep with your infant, they're going to die. Basically this is a way for your infant to die and the risks are very high and you should never do this.

Ted Roosevelt V:

Because you're going to roll over on them, or because...?

Emily Oster:

Because you're going to roll over on them. Exactly.

Ted Roosevelt V:

Okay.

Emily Oster:

The actual data on co-sleeping does suggest there are some elevated risks of SIDS or of suffocation, but they are quite small and they're sort of within the sets of risks that people might take. For some people that is the only way people can get sleep, and so there are some very clear trade-offs. If instead of saying "co-sleeping with your baby will make them die," we would say, "there's a small elevated risk from co-sleeping. If you are going to do it, here are the safest ways to do it. And by the way, never fall asleep with your baby on a sofa because that is really dangerous." That messaging, while not as simple, would help people make better decisions. The goal can sometimes feel like, our goal is to get people to do what we say. And I don't think that be our goal.

Ted Roosevelt V:

So I'm curious if you think about the messaging... where the message comes from, maybe—I'm wondering if that impacts how you think about it, because if it's a doctor to a patient, it can be a much more nuanced conversation. If it's Fauci to the country in a moment of COVID-19 and I think there are very credible critiques of how things were handled, but it is a very different environment when you're trying to message en mass versus message individually.

Emily Oster:

Absolutely, and you will often hear people say, well, in a situation like Fauci, the only way to do it is to just message something super clear. The problem is if you then turn out to be wrong, you have lost people's trust and you cannot get it back. There's a bunch of examples of this in COVID. I think the clearest one is masks. People were told "Don't have masks, nobody needs masks. Masks are stupid." And then five minutes later, they were like, "masks are amazing. Everyone has to wear them all the time and if you don't wear them, you want to kill people." And nobody ever said, "we're sorry, we were wrong." It would've been much better upfront. And same with, do vaccines prevent transmission? There were all kinds of places where the uncertainty in what we thought was known, and I think the long run outcomes would have been better if they had been more willing to communicate the way we communicate about wildfires. Here's where the fire is now, here's where we expect it to go tomorrow. I'm going to show up tomorrow and tell you what we've learned and what's going to happen next. If we had said, here's what we think now about masks or here's what we think now about transmission and we're still learning more and we're going to come back and tell you when we know more, I think that people would had more trust in that messaging.

Ted Roosevelt V:

Yeah. I'm not like a big defender of how things were handled with COVID-19 at all, but it does bring up another area I'm very curious about, which is the role of data in messaging at all. Because what I keep finding, and I spend a lot of time in the climate change space, is that whether it's climate or COVID-19 or vaccines that one's opinion about that is really more aligned to group identification than it is to the underlying facts. And I can see you sort of nodding along here. And I wonder how much do you think data is useful from a messaging standpoint to people versus people are going to find the data that they want anyway.

Emily Oster:

In many of these spaces there are people who are very committed in one direction or the other, and then there are people who are in the middle. And when I think about messaging, I'm often—about data, I'm often thinking about the people in the middle. So in the case of something like vaccines, there are a lot of people who are like: I haven't read anything about vaccines, but I'm really very—I know the basic structure. This sounds good to me, it's fine. I get the vaccines, I'm moving on. And then there are a set of people who are very committed to the position that vaccines are bad in some way and that they're problematic. And I think for many of those people, there's no data you could show them. I mean, I think the reality is: RFK Jr. believes that vaccines cause autism. I'm not sure why. I'm not sure what piece of evidence or where that has come from, but I think that he really believes that and I don't think there's any data you could show him.

Ted Roosevelt V:

Isn't there one kind of debunked paper that came out from England or something like years ago?

Emily Oster:

There's one debunked paper, yes. There's one— like, I know where the idea comes from, but given all of the evidence countering the idea, I'm not sure where the persistence of the belief comes from. But where I think in that case there are a lot, and actually a growing number of people, is in this place in the middle where it's like, "I thought these were good, but I'm hearing that they're not good." And that's actually a place where I believe that meeting people with data can be very helpful and helping them understand: somebody cited this statistic, where did it come from? I think the value there of being able to say, "okay, I located the piece of evidence that he is, I think, pulling on and here is why actually that wasn't a coverup and here is why there are some issues with the way that study was run and here is why in fact there's much better data, which is the thing you should be listening to"—I think that's valuable for those people in the middle who might otherwise be pulled one direction or the other. And in many cases, of which vaccines are one, actually, that's a pretty big share, and if all of those people got vaccinated, got their kids vaccinated from measles, we are in a very different situation than if none of them do. And so I think rather than sort of despairing that one cannot get the edges, there's a scope for the middle where I do believe that data can help.

Ted Roosevelt V:

And I just wonder how much of the problem do you see as being a classic systematic problem of social media, the systematic problem of how we absorb information these days, particularly at moments of medical vulnerability where fear is going to be heightened?

Emily Oster:

Yeah, I think that's a huge share of it. I mean, I'm an economist. Most of the decisions that are made are based on somebody's incentives. And so you think about: if I'm writing a media story and I'm writing the headline, I want a headline that people click on. If I'm writing the story, I want a story that people click on, so I write the story in a way that surfaces a headline that people click on and it goes down. If I'm thinking about even research that I want to do or research that journals want to publish—you know, journals like to be in the media also, and that's going to pull them towards studies where maybe they're thinking about the kind of "clickiness" of the study rather than the quality of the findings. That incentive sort of rolls around and maybe it's ultimately our fault for clicking. It's you. It's you that is doing this, Ted, too much clicking.

Ted Roosevelt V:

I knew it came back to me some way or another. You got very much into the COVID-19 conversation—which was just a conversation that I don't think anybody has entered and it hasn't been fraught for them to some degree—and there was a reasonable backlash in that conversation. It didn't seem to bother you much that there was a backlash and I'm wondering why or how you're able to keep a distance from that in those conversations.

Emily Oster:

So—it really bothered me.

Ted Roosevelt V:

Oh, it did? Okay. You handled it well.

Emily Oster:

There were a few different moments during COVID in which I made sort of statements out in the world that then produced a backlash. Probably the most...the earliest of which was sort of saying that we should reopen schools, and I spent a lot of time advocating for that. And definitely there was a lot of backlash and nobody likes being told that they—not even me—that they're killing people and that they're a terrible person who's... whatever. I think more seriously in those conversations, I was very sure that what I was saying was important and correct, and I think particularly around schools, I think I was proven that I was right. We should have opened schools more and I don't think it was a good idea. And I think many people now who at the time said, you're a terrible person now, would say, well, actually that was a very big mistake that we made.

Ted Roosevelt V:

Yeah. The idea of entering the public square on something that's so fraught, like COVID-19 and the sort of emotional toll that something like that can take on you, I think that's hugely important for people, that sort of substantive people to step up and make a point. Do you get to a place where you're like, "I did the right thing. I feel good about it. I'm glad I stepped into that mix"?

Emily Oster:

I felt bad that people were yelling at me because it's frustrating to have people yell at you and feel like you're not listening to me. Did you read the whole piece? Did you read all the words? Because I was very nuanced. You didn't read all the words. Those things were frustrating and made me upset, but I was glad that I did at the time and I'm still glad that I did it. If a bunch of people yelling at me online meant that some poor kids got to go to in-person schooling and learn some things, that's such an amazing trade. I'd take that a million times over

Ted Roosevelt V:

Emily. I was going to say: you just ran the Boston Marathon this year.

Emily Oster:

I did. Yeah.

Ted Roosevelt V:

I am so profoundly impressed. I've run a number of marathons, but the idea that you've run two marathons...the first one you qualified for the Boston Marathon is insane because nobody runs a good marathon their first marathon. And that you then ran a 3:15. Are you a natural runner? Have you been running before or were you, like, on the couch and you're like, maybe I'll run a marathon?

Emily Oster:

Yeah, no, I've been running since I was a kid, although I got super into it during the pandemic. It's very important to my mental health, I guess.

Ted Roosevelt V:

I totally agree. I heard you on the podcast, Allie On The Run, and she raised the all important topic of how efficient the bathroom lines are during races. Anyone who's been in a race knows that this is a major issue, and so it made me actually wonder: are there other areas of society outside of public health where you think you could use more economists doing the kind of practical, real-world work that you do?

Emily Oster:

There is a lot of waste in government. I mean, it's a sort of interesting... it's an interesting space because on the one hand, I think that many of the things that the Department of Government Efficiency was trying to do were cruel and did not actually result in any good outcomes, but I think not dissimilar to the porta-potty lines, there are a bunch of places where it feels like the public sector has some low-hanging fruit relative to some private sector operations. I'm not sure that economists are the people so much to fix that. It's more like systems engineers, but certainly there are places that I've interacted where it feels like there's some efficiency gains to be made here. If we thought about this with concrete specific outcomes in a way that one does inside a business.

Ted Roosevelt V:

Well, I can confirm porta-potties need more work.

Emily Oster:

Yeah, it's a very difficult situation with the lines. They're just not optim— I mean, that's an easy one.

Ted Roosevelt V:

It's such low hanging fruit and—

Emily Oster:

We should put Elon. Elon should be in charge.

Ted Roosevelt V:

Oh my God. I would love it. I would definitely—I don't know if he would think it was the highest and best use of his time, but I would love him to do it.

Emily Oster:

I know.

Ted Roosevelt V:

You had a very popular podcast that you ended, I think in March, right?

Emily Oster:

Yes.

Ted Roosevelt V:

Why did you stop the podcast?

Emily Oster:

Well, we could come back in some form. Time is finite and there are ways that I think I can be most...maybe it gets back to your last question, which is trying to think about what are the things that I'm doing that are most impactful and most unusual, and I think that the writing that I do is the most impactful stuff that I'm doing, and I wanted to be able to spend more time on that. And I wanted to rethink, what's the right focus of the podcast and what's the right way to select topics. I dunno, I needed a break. Podcasting is, I mean, you know this, like it's a... it's hard. It's hard.

Ted Roosevelt V:

Are you suggesting podcasting is not unique and unusual in the world?

Emily Oster:

No. My podcast was not unique and unusual

Ted Roosevelt V:

No? [laughter] I'm kidding.

Emily Oster:

No comments.

Ted Roosevelt V:

So where are you focused today? What is the sort of tip of the spear for you?

Emily Oster:

So we're building out ParentData, which is really an extension of the books and an extension of the Substack newsletter, but I want it to be the place that people who are trying to conceive through "my kid is five," come for their middle of the night answers, come for their big picture context, come when they see something panicky on the internet and they want to understand more about it. I am pretty focused on trying to understand how I can raise awareness about that resource, how we can build it to make sure that it is usable as possible. And it's an opportunity to learn in a totally different way, but to learn about what does it take to run a business successfully and to help people change their mindset, to be able to process the information that they're getting in a way that is more real time for them. I want to have the answers for people in the middle of the night, but I also want to have taught people to fish a little bit around this kind of more panic-inducing news stuff. You have the ability to just take one more moment to sort of look for some more information and then be less panicked.

Ted Roosevelt V:

I mean, I hear you saying you'd like to see more critical thinking in the world.

Emily Oster:

I would like to see more critical thinking and more data literacy. More critical thinking and data literacy, that would be so great.

Ted Roosevelt V:

I've also heard you talk about the idea that you like being approached in person with these questions. I don't think of economists as being extroverts and wanting to have their information shared on a one-to-one basis, but you get something out of that, it sounds like.

Emily Oster:

Yeah. I mean, I like to help people and I'm very interested in people's parenting and pregnancy situations. I like when people ask me questions in person. I don't mind. People who approach me on the street a hundred percent of the time know me from Instagram (because otherwise how would they know what I look like). They think of me sometimes as someone who's been there and has talked to them right in their phone for three years while they're running through their infertility journey. It's not like telling some random stranger, it's like this person was there with me all of these years and now I want to tell her because I'm excited, and that's such an amazing privilege.

Emily Oster:

One of the things that has happened is people will often recognize me in airports and sometimes when I'm with my family. And so, the other day we were in the airport and we were in the passport line and the person in front of us, she went up to the window with her husband and she turned at the last minute and she was like, I just want to say, I love your stuff. We're on our first baby. I was like, oh, congratulations, great. Then her husband—he's asking her, he's like, who is that? She's like, oh, she's the person who wrote Expecting Better, and he just turned around to us and he was like, "Clutch." My husband was like, that was the best. That was the best one.

Ted Roosevelt V:

Question we ask everybody, and I'm going to be really interested by your answer, is: what does it mean to be a good citizen?

Emily Oster:

I think it means to give back what you can. To recognize, what is the capacity that you have to contribute and to try to live up to that capacity.

Ted Roosevelt V:

And I think you're doing a phenomenal job at it.

Emily Oster:

Thank you.

Ted Roosevelt V:

Thank you, Emily, for joining us today. I really appreciated it.

Emily Oster:

Well, this was awesome, Ted. This was really, really fun. I had a great time.

Ted Roosevelt V:

That was such a great discussion. Thank you so much, Emily. I really truly appreciate that you are using your great skills and talents in the service of the public. Listeners, please check out parentdata.org. It is packed with insights and a rich source of information. And Emily's latest book is The Unexpected: Navigating Pregnancy During and After Complications.

Ted Roosevelt V:

For more great interviews like this one, please make sure you're subscribed to the podcast and consider leaving us a review. Good Citizen is produced by the Theodore Roosevelt Presidential Library in collaboration with the Future of StoryTelling and Charts & Leisure. You can learn more about TR's upcoming presidential library at trlibrary.com.