Emily Oster on data, pregnancy and parenting

Speaker Key:

EO              Emily Oster

AL              Andrew Leigh


EO              In general when you are pregnant, the first thing that happens is people are like, okay, here are the list of things that you can’t do and the list of things that you should do. I am a person who likes to ask why.

AL               My name’s Andrew Leigh, and welcome to The Good Life, a politics-free podcast about living a happy, healthy and ethical life. In this podcast, we seek out wise men and women who have lessons to teach us about living life to the full, with humour, pleasure, meaning and love. We’ll chat with musicians and athletes, CEOs and carers about making the most of this one precious life.

                   If you like this podcast, please take a moment to tell your friends or rate us on Apple Podcasts. Now, sit back and enjoy the conversation.

                   Emily Oster is an economics professor at Brown University. I first met her when she was completing her PhD at Harvard, where she graduated in 2006, before taking a job at the University of Chicago. Emily’s research interests are extraordinarily broad, from the impact of television on gender equality in India, to how people change their eating patterns after being diagnosed with diabetes.

                   But in the public arena, she’s best known for two books, Expecting Better and Cribsheet. Both of which discuss a data-driven approach to decision-making in pregnancy and parenting. It’s these books we’ll spend most of our time discussing today. Both for their wisdom, and for the wise way that Emily goes about gathering evidence to make better decisions. Emily, welcome to The Good Life podcast.

EO              Thank you for having me. It’s nice to talk to you.

AL               So, your parents were economists, but not just any economists. Your mum, Sharon Oster, works on business strategy and non-profit organisation management. The first woman to receive tenure at the Yale School of Management. Your dad, Ray Fair’s a macroeconomist, but my favourite bit of his research is his analysis of the rate at which runners slow down as they age, which he's just updated last year. What was it like to grow up with economist parents, and really good economists?

EO              Yes, it’s interesting. I think when you are inside your family, you think your family is regular and normal. Then sometimes when, then, you step out of it you’re like, wow there were some unusual features.

                   I think, for me, the one thing was in… There’s a lot of use of economics in everyday life in my childhood. So, the answer to the question, why don’t we go grocery shopping at the grocery store, why do you order groceries, was, oh, well my opportunity cost of time is very high, as opposed to… Which I recall being like, oh okay, of course, that’s an excellent explanation. But when you say that to other people they’re like, wow your family must have been really weird. But they were very good parents.

AL               Did you always want to be an economist?

EO              No. I actually… I was always very interested in research, but for a long time I thought I would do something more science-oriented and then at some point in college I realised that the family business, as we say, was calling.

AL               On the topic of parenting research, you were a subject of a rather interesting experiment by Katherine Nelson when you were two years old. Can you tell us a little bit about that?

EO              Yes. So, at some point when I was just shy of two, my parents were at a dinner party, I guess that you have to go to if you’re a professor, with some other professors. One of the wives of one of the economists there was a child development researcher, and she was interested in kid’s vocalisations in the crib, basically when kids talk to themselves, and that was something that I did a lot as a kid.

So, my parents offered to audio tape, use a tape recorder, to record all of the stuff I was saying to myself in the crib. And ultimately this researcher and some other people that she worked with ended up writing a book that’s all about the language development of a little kid using these tapes of me. It’s not that interesting. I don’t find the book that interesting. Although I like the parts that are about me, I’ve never managed to get quite into the weeds of linguistic development.

AL               And it also seems to me that the book, Narratives from the Crib, which was reprinted recently with a foreword by you, is almost the antithesis to the way you approach research. To draw broad conclusions from a sample size of one seems the opposite of what I think of as Oster-nomics.

EO              Yes, although I will say that I think in the particular case of what they’re trying to do. I can see the value in having a huge amount of data on one person, because they’re very interested in the order with which things are done, linguistically, as opposed to… I think it would be a particularly weird way to draw conclusions about what is the age at which children learn to do X because, of course, you don’t know if I’m the typical kid on this or not.

                   But some of these things, like does the past appear before the future in the language, I can see why you’d want to go deep into one observation. But it’s certainly not the same way that I do most of my research.

AL               Let’s dive into Expecting Better, your 2013 book, subtitled, Why the Conventional Pregnancy Wisdom Is Wrong, And What You Really Need to Know. I’ve got to say, I love the book. I have four copies just sitting in my cupboard, waiting for one of my data-driven friends to tell me they’ve fallen pregnant for the first time and so I can drop one of these books into their hands.

                   You said you were prompted to write the book because of a frustration with the way the literature was as you read it when you fell pregnant yourself. What was that frustration?

EO              I think in general, when you are pregnant, the first thing that happens is people are like, okay, here are the list of things that you can’t do and the list of things that you should do. I am a person who likes to ask why, and particularly in this case, where there actually… You would get different recommendations from different sources. So, a question like, can I have a cup of coffee? Some people would say you can have as much as you want, some of these sources said you cannot have any.

                   I was like, okay, can you tell me why? And there wasn’t any source where I could really get the answer to, why are the recommendations what they are? Why are they sometimes disagreeing? What is the reason to be concerned? So, in a lot of these questions, I started to spend a bunch of time really trying to dive in the literature and it turns out, a lot of the issues that come up in those analyses are the same issues that come up in the kinds of work I do in economics. So, my training was, in some ways, pretty well-suited to thinking about a lot of these issues.

AL               I think coffee is my favourite example in the book, because you talk there about what we economists refer to as selection effects, and not just the socioeconomic effects, but the nausea effects. Can you say a little bit more about why the observational studies might have been giving us the wrong answer?

EO              Yes, so I actually share your view. I think a lot of economists like this, because it’s in some ways the most interesting methodological problem. So, the issue is with coffee, we’re concerned about the risk of miscarriage, and whether coffee consumption increases that risk. There’s basically two problems. One is that there are other things that increase the risk of miscarriage. A notable one is maternal age, but also smoking. And women who drink coffee, in general, are both more likely to smoke and are, on average, older.

                   So, when you’re just comparing people who drink coffee to those who don’t, you have this first problem that they’re different in these other ways, and those are associated with the coffee drinking and with miscarriage. And that’s a problem you have in a lot of these places.

                   In the particular case of coffee, there is an additional problem which is that women who are more nauseous are less likely to drink coffee, partly because coffee is just something that is not something you want to drink when you’re not feeling good. But nausea also is associated with a healthy pregnancy. I mean, it’s okay if you’re not nauseous, but on average, nausea is a good sign. That means that, basically, unless you can also see really detailed information about people’s nausea, you’re going to mistakenly attribute miscarriage to coffee rather than to nausea.

                   One of the ways I say it in the book is, if you feel good enough to have eight cups of coffee a day, that is not a good sign independent of the coffee and it’s actually very hard to learn about whether coffee is particularly bad or not. But I think ultimately the data suggests that, certainly, some coffee is fine and whether you want to draw the line at six cups a day is more ambiguous. But, since that’s well outside of what most people are thinking about it’s probably a little bit less relevant.

AL               And you also talk there, too, about the notion that perhaps there’s not a single right answer. And maybe it might turn on how much you really love coffee and trading off very small risks, or uncertain risks to the baby against how happy mum will be to have the right amount of coffee.

EO              Yes, and I think this comes up, this issue of trading off preferences and accepting that it might be worth it to take some risks because these risks may be very small relative to the benefits to the parents. This comes up in both books, that I feel like a lot of our discussions around pregnancy and parenting assume that the parents have no preferences and that there’s nothing that they want. Sometimes we need to recognise that that’s not always the case, that we should put some value on everyone in the family, not just the baby.

AL               Moving onto alcohol, for which, I think, you note that a number of the key studies are Australian. What does the research evidence tell us about drinking while you’re pregnant?

EO              The first thing to say here is that the first thing that the research says is that drinking a lot while you’re pregnant, binge-drinking, is really dangerous and is associated with all kinds of negative outcomes. But what I try to do in the book is think about the question of occasional drinking and what does the evidence say about having, say, one drink every… Few drinks a week during the second and third trimesters.

                   And there I think that when we look at the best evidence, some of which does come from Australia, it does not look like there are negative impacts of occasional drinking on the baby. Again, this is a place where I think some people look at that evidence and say, you know what, it’s just not worth it for me, I’m not comfortable with that. And some people will say, okay maybe I’ll have a couple of glasses of wine on a special occasion because I think that’s something that would… That would be within what I’m willing to do.

                   So, I think it’s useful there to look at the data. And I think partly because at least in the US, a lot of obstetricians will tell their patients, oh it’s okay to have a little bit. But since the official recommendations are so stringent that you shouldn’t have any, I think people often find themselves again in this place of, well why do you say that it’s okay to have a little bit. Why is that at odds with the official recommendations?

                   I think it’s useful to say, here’s what the evidence says, here’s why there might be these different recommendations and you can then decide for yourself what is going to be the choice for you.

AL               You’ve been criticised for this. You’ve got a few critics for Expecting Better. But this recommendation, in particular, has been criticised. One of the arguments that gets made is that we shouldn’t be telling pregnant women that some drinking is okay because of the risk that some drinking will turn into a lot of drinking. What’s your view about that?

EO              I think, first, the question of what should policy be as opposed to what does the data say is actually not that obvious. I think that there are considerations we have when we make official policies that go beyond the question of the data. Unfortunately for me I’m not in charge of making official policies, but I am a little bit uncomfortable with this idea…

AL               My party just lost an election in Australia, so we are not responsible for…

EO              But I am a little bit uncomfortable with this idea that, somehow, women cannot be trusted with their choices and that we should hide information because it will be misused. I think that that’s, in some ways, it’s not respectful.

AL               Tobacco. Do you find the same sort of thing there? Are you concerned about the official recommendations, or do you go with the official body on [overtalking] that?

EO              No, I think on the tobacco we go with the official body that smoking during pregnancy, or actually outside of pregnancy, is not good. If women can quit, of course it’s quite difficult to quit smoking, but if you are smoking, during pregnancy is a good time to try. There are, actually, pretty strongly demonstrated effects on birth weight and there actually are some negative effects on the kids even after the baby is born, so pregnancy is a good push to quit smoking.

AL               What about the huge list of foods that pregnant women are told to avoid. Soft cheeses?

EO              Here, I spend a lot of time trying to help people understand why we have these restrictions on foods. Soft cheeses are one. Unpasteurised soft cheese can carry listeria. It’s not super common but I think that it is something that I avoided, not all soft cheese but soft cheese that wasn’t pasteurised.

                   But then some of the things you hear like all deli meats should be microwaved or, maybe you guys don’t have that one, but sushi should be avoided. These are not things that are particularly well supported in the data. There isn’t really any reason to avoid sushi. You should try not to eat spoiled sushi but that’s something you should be doing when you’re not pregnant also.

AL               And deli meats. I think you come down on the view that you avoided turkey but didn’t worry too much about other deli meats.

EO              Exactly.

AL               What about fish? You talk about the mercury versus omega-3 trade-off. What are the implications of that?

EO              I think a lot of people… You hear about fish on the one hand, that fish is good for brain development, on the other hand it may have mercury and I try to… Those things are both true, but of course different fish have different amounts of this.

                   There are some kinds of fish like shark or swordfish which have a lot of mercury because the fish is very big and relatively limited omega-3s, so that’s probably not the best fish choice. But then there are some little fishes like herring or sardines where they actually have a lot of omega-3s and almost no mercury and so those are better choices.

                   I think, in the end, a lot of women end up thinking, oh, you know what, I’m not going to have any fish at all, which is actually not a particularly good decision, because even if you had a little bit of some of these high mercury fish that’s not that much mercury. We should not be afraid of fish. Fish is good for you.

AL               And then you had this interesting discussion around painkillers where you, to my surprise, talked about quite different evidence around acetaminophen, ibuprofen, and aspirin. If you’ve got a headache and you’re pregnant, which of those should you go for and which should you avoid?

EO              Generally, the advice is to take acetaminophen and I think that the evidence on… There are some negative aspects of aspirin in particular in pregnancy. Actually, we now hear some people nosing around the edge that acetaminophen might also be not good, but I think that that evidence is pretty poor. So, if you have a headache, if you feel bad, Tylenol is fine.

AL               And then there’s things to do. Dying your hair.

EO              Yes.

AL               Gardening.

EO              No.

AL               Are these things to worry about?

EO              I think dying your hair is fine. There was some research on hairdressers which suggested that they were more likely to have birth complications. But I think it turned out that that evidence is almost certainly driven by differences across women, not by them hairdressing or the hair dye specifically.

                   But gardening is a little more complicated because it is a risk factor for toxoplasmosis, which can be dangerous for the baby. So, although that’s not something people talk about much, that probably is something to be a little more cautious about during pregnancy.

AL               This is a soil-borne parasite?

EO              It is a parasite which can live in the soil, yes. Particularly if there are… It’s often carried by cats, carried by other animals but it is often carried by cats, and it ends up in the soil because of cat poop, basically.

AL               I thought it was interesting where you ended up concluding that actually gardening is a higher risk for toxoplasmosis than emptying the cat litter box.

EO              Yes. Although if you want your spouse to empty the cat litter box you should just make them do that.

AL               Yes, well, you have a right to do that if you’re carrying their child. What about exercise? What are your recommendations there?

EO              Exercise is good. If you can exercise, there is no reason not to. Even, actually, at a pretty high level. There’s some discussion of, if you’re an Olympic runner, during the period in which you’re running there’s somewhat less blood flow to the baby but that doesn’t seem to be a problem. In general, exercise, particularly… Well, not particularly but things like yoga or… There’s a bunch of kinds of exercises that can be very effective and I think that staying at least vaguely in shape is helpful for, say, childbirth.

AL               Is airplane travel a risk?

EO              No. Not at the levels that most of us do. People think a lot about radiation but the amount of radiation you get… Unless you’re flying back and forth between you and me many times, you’re not going to get the kind of radiation we’d be worried about.

AL               Sleeping on your back?

EO              This is a place where we keep revisiting it, because the data is really evolving. Women are told not to sleep on their back because of the risk of stillbirth, and I think that in some ways the jury is still out on whether that’s true. Most women do not find it especially comfortable to sleep on their back so it may not come up that frequently. But I think probably there sleeping on your side is a better choice, at least until we know more.

AL               What about the recommendation of taking bedrest to prevent pre-term labour?

EO              That is not a good recommendation. This is a place where the evidence is very clear that there is not any condition for which bedrest is a good choice, but it is still prescribed, at least in the US, quite frequently. It’s something where I think we need to push back a little bit. Because it not only is not good at preventing pre-term labour, but it also has other negative consequences, like you lose time with your family or time at your job and also there can be some… It can have some impacts on your muscles.

AL               Yes, I think about the findings around elderly people and falls. Most of the risk of falls seems to be the subsequent need to have bedrest and the complications that causes. Presumably, if we think that spending a lot of time in bed is bad for elderly people, it’s not obvious why we would think it was good for pregnant women.

EO              Absolutely.

AL               As you wrap up Expecting Better, you talk about labour itself. What are your views on the controversial topics of home births and epidurals?

EO              So, with the epidural I think that the evidence suggests that maybe it slightly complicates some aspects of labour. Maybe it makes it a bit longer and has some… There are some interventions that will have to happen if you have an epidural that wouldn’t happen if you did not. On the other hand, it’s very effective pain relief so I think that’s a trade-off that people can think about. Home birth…

AL               How much longer?

EO              Not very… 15 minutes. Which could be very bad in the moment, but all of these things are very uncertain. Some people… It’s not like if you get an epidural for sure you will be 15 minutes longer. For some people, it speeds up labour. So, there’s a lot of differences across women even in the way that they respond to that treatment.

                   The question of home birth is more complicated probably because it differs so much across places. So, in the US, it does look like home births are a bit more risky. But that largely seems to be that a lot of home births are with not-certified midwives, just people who are not especially carefully trained. So that’s a mistake. But the question of, what about a home birth with a trained midwife in a setting in which this is more common? I think the risks there will be smaller.

AL               Let’s move now to your recent book, Cribsheet, which last time I checked was sitting atop the Washington Post’s list of bestsellers. Congratulations.

EO              Thank you.

AL               You, again, run through a wonderful raft of evidence, starting off with some of the first decisions that new parents have to make. If it’s a boy, should the child be circumcised?

EO              Much of what Cribsheet is about is in some ways just saying, hey, you’ve got to choose for yourself, your preferences are important. Circumcision is a good first example of this, where there are some small risks, very small risks, like infection. And then there are some very small benefits, like a lower risk of penile cancer, which is an extremely rare cancer.

                   So, ultimately the question of whether you will do this or not should lie with what do you think is going to work for your family. And people have all kinds of preferences for this, for religious reasons, or because they want their kid’s penis to look like his dad’s penis. These are all kinds of reasonable considerations that you should make. There isn’t something in the data that’s going to push you one way or the other about doing it.

AL               What about swaddling to get the baby to sleep well? Do we have good quality evidence on swaddling?

EO              Yes, I love the evidence on swaddling. So, the idea with swaddling is that it can help the baby sleep.

                   This is actually a place where we have the opportunity to study it pretty carefully because you can see a baby, the same baby, sleeping swaddled and not swaddled. So, you don’t have to rely on things like asking parents, how does your baby sleep and comparing them.

                   There’s at least one very nice study where these researchers took these babies into the lab. They swaddled them up in this swaddle blanket that contains all these sensors on it, it’s like this weird robot baby, and then they video tape them. So they have all this information and what they find is, basically when babies are sleeping, at various times they will start to rouse a little bit. Then in the swaddle babies, those initial rousings are less likely to lead to further rousings and then awakeness.

                   Whereas in the un-swaddled baby, the same baby, if they’re not swaddled, when they start fussing around a bit they’re more likely to end up crying. This is mechanistic evidence, I guess you would say, for why swaddling helps babies sleep longer. So swaddling is a good idea. Except eventually you’ll have to stop swaddling them, which is difficult.

AL               One of the things with three little boys that I most enjoy, is this notion that maybe going off and playing in the dirt could be good for them. Am I grounded in good science there or am I really just being a lax parent?

EO              I think you’re mostly grounded in good science. This is related to this idea that playing in the dirt will raise your kid’s immunity and will have other good impacts later. I think we increasingly have evidence that these kind of immune challenges or these exposures are good for longer-term health.

                   This is called the hygiene hypothesis. The idea is that people are too clean and we should be less clean and that that would be better for our long-term health. I certainly let my kids play in the dirt. I’m not too neurotic about this.

AL               This is something where you talk later in the book about the early exposure to things like nuts. You say, actually, maybe we’ve got pretty good evidence that ensuring that kids don’t touch nuts early on could be increasing the rate of allergies.

EO              Yes, I’m still curious.

AL               Tell me a little bit more about that.

EO              In Australia, what is the recommendation about allergen exposure early on?

AL               Relatively similar to the US, so we’ve seen… We don’t have… It’s not like Israel where there is some standard, peanut-based snack. Schools are largely nut-free and recommendations are around avoiding contact with nuts early on. We’re seeing rising nut allergies. We don’t have a Gideon Lack here yet.

EO              Yes. Not yet. In the US this has typically been the recommendation that people should avoid exposing their kids to nuts, peanuts in particular, but also all kinds of allergens because the idea was that would increase the risk of allergies.

                   But actually, within the last few years we have quite good evidence from a randomised trial that exposing kids to nuts early, and by early I mean four months, basically as soon as they are able to have solid food. Exposing them to nuts decreases their risk of being allergic.

                   The effects here are very big. This is a trial in which they took kids and half of them were exposed to peanuts early and half of them were told to wait until a year or something. The difference in… There’s about a 70% reduction in the risk of peanut allergies. This is a pretty high-risk group, so something like 17% of the kids who were not exposed to peanuts ended up with a peanut allergy versus only 3% of the kids who were exposed. You rarely see effect sizes that large in data.

AL               So you could get rid of three quarters of peanut allergies through early exposure.

EO              Exactly. Yes.

AL               Wow.

EO              And in the US there’s of course now a bunch of start-ups that will help you with how to expose your kids to peanuts through powders and stuff.

AL               We’re starting to turn the temperature up to get into more controversial issues, so let’s dive straight into one of the most controversial, breastfeeding. What is the evidence on this?

EO              So in the US you’re told that breastfeeding is the most important thing you could possibly to and has all kinds of effects on you and your baby and better friendships and all kinds of other things. I tried to dig into what is actually supported in the best data, and I do find that there are some benefits that are supported there, in particular a lower risk of some kinds of rashes in the first year. Lower risk of diarrhoea, again in early life. And actually, maybe some reductions in breast cancer risk for the mom. But some of…

AL               Quite a significant reduction.

EO              Quite a significant… Yes.

AL               I think about a quarter of…

EO              It’s like 25%. So those are some good reasons to do this.

                   But some of these claims about say, the impacts of breastfeeding on IQ or obesity or long-term health for your kid. Those don’t seem to be supported in the best data. So, I think the picture is maybe a little bit more nuanced than you might get. I think particularly because a lot of women feel a lot of shame if breastfeeding doesn’t work for them.

                   I think that seeing the data and saying, yes this is a good thing to try and we should be supportive, and, here, it has these benefits. But if you can’t do it, it doesn’t mean your kid’s going to be fat and stupid because you couldn’t do this. I think that is an important message.

AL               You’re relying here on quite different data than many people who write about breastfeeding. What is it about the Belarus randomised trial that marks it out as being different from the observational studies we have?

EO              Yes. Most of the data we have on breastfeeding is from these observational studies where you compare the children of women who breastfeed to the children of women who do not. That is subject to why these concerns we mentioned in, for example, the case of coffee, that the kind of women who breastfeed tend to be better educated, richer, than those who do not, and that can bias our results.

                   I spent a lot of time relying on two kinds of data. One is that there’s a large, randomised trial that was run in Belarus in the 1990s in which some women were encouraged to breastfeed, and some were not. We can use that, because that encouragement is random, we can use that study to draw more stronger causal conclusions.

                   The other thing I do is, within these observational studies, I try to be more careful about which of them I give more weight to. In particular, some of these studies are like comparing siblings, and some of them have much better sets of controls and I rely more on those than on some of the studies which are able to do less adjustments for differences across moms.

AL               And you talk also about the one of the other big controversies around sleep-training, the notion of cry-it-out sleeping, and whether or not that is effective or whether perhaps it scars children in the long-term. What do you find about the efficacy of getting kids just to cry it out as a way of learning to sleep on their own?

EO              So, the first thing is that letting kids cry it out will on average improve their sleep and will on average improve the sleep for the parents, and also has some pretty good impacts on parent’s… On depression and marital satisfaction. There actually are a bunch of reasons why these cry-it-out methods and improving kids sleep are good for the family overall.

                   People, of course, worry that this is going to scar their kid forever and that they’re going to not ever be attached to them and they’re going to hate them. I will say as someone who has let some kids cry it out, definitely is very difficult to do in the moment. But there isn’t any evidence to suggest that this is bad for your kid.

                   In the long-term, we have studies which have followed kids until they’re five or six. Kids who cry it out look just like the kids who did not. If anything, in the short run, it looks like the kids are happier, although that may just be reflecting what their parents say about whether they’re happy because the parents are happier.

                   It doesn’t mean that this is for everybody. Some people will just… This is like… I’m not… They don’t want to do this. The kid is sleeping with them. There’s various reasons you might not want to do this. But if it is something that you think would work for you, or if you’re really struggling on sleep stuff, this is worth trying.

AL               You have what I think is one of the most biting lines in the book. You say the argument against sleep training is theoretical. I can almost hear you spitting the word out there.

EO              Theoretical.

AL               From an empiricist point of view.

EO              It’s theoretical, yes. People say, well this could be bad. Sure, anything could be bad. It could be good. That’s the problem with theory. You need data.

AL               Yes. Sleeping positions. Dr Spock in the 1980s was telling us kids had to be on their stomachs. Now, we say kids have to be on their backs. And there’s a similar conversation going on as to whether co-sleeping is acceptable or not. What did you find when you looked at the evidence as to whether it’s all right to have the kid in the bed?

                   I remember we never did it for our number one and we did it all the time for our number three, which either says something about our experience as parents or else just how exhausted you are when you’ve got to your third.

EO              Yes. I cannot imagine. The co-sleeping stuff is complicated, because I think we see very strong views on both sides. With some people saying, this is the most natural way to sleep with your kid and everybody should do it. Then on the other side, people say, this is incredibly dangerous and you definitely should never do it.

                   I think the evidence is really, truthfully in the middle of those. One thing is there are some more and less safe ways to do this. We talked about smoking in pregnancy. This is a place this comes up again. If you’re smoking, if your partner’s smoking, if you’re drinking heavily before bed, if you have a lot of covers and stuff in the bed, that makes co-sleeping dangerous. And in those circumstances, it is actually, probably not a good idea to co-sleep.

                   But if you’re going to do this in the safest way, so not drinking heavily, not smoking in a bed with less stuff, there is still a little bit of risk. But it is very small relative to some of the other risks, like driving a car that you’re probably taking every day without thinking about it.

                   So, I think that does mean that it’s… There’s a component of this which is parental choice and is what is going to work for your family. I am guessing many people co-sleep more with their later children because they’re… That’s the way your family can get some sleep, and you need to be rested because, say you have other children around or you just need to be doing stuff.

AL               And you have a really interesting observation in there about something that I’d never thought of as being such a big risk, sofa sleeping. Say a bit more about that.

EO              Yes. Sleeping on a sofa is probably the most dangerous sleep environment because if you fall asleep sitting up with a kid on the sofa it’s easy to slide down and because there’s a lot of cushions and soft stuff those are very risky. I think part of what’s very tragic about this in some cases is that people will sometimes sleep with their kid on the sofa because they’re trying to avoid sleeping with them in the bed. So, they’re trying to stay up and sit up and then there can be tragic consequences of that.

                   I think that that is both something to avoid but also is another caution about being careful about the messages that we send. There are places where we should be sending a more nuanced message, like, if this is the only way you can sleep, please do it as safely as possible. As opposed to, under no circumstances should you do this, and then leading to people doing things that are even less safe.

AL               What about other activities on the sofa. If you have the TV on with Baby Einstein going there, are you going to turn your child into the next Einstein?

EO              Unfortunately, Baby Einstein is not the way to turn your child into the next Einstein.

AL               Who knew?

EO              Babies cannot learn from videos. The research on this is kind of funny, because they will give some parents these videos and then ask them, did your kid learn words from this? And parents are like, oh yes, my kid definitely learned a lot of words from this. It was so great. But then it turns out that actually, when they look at them, all kids learn words, even the kids who don’t watch Baby Einstein. So, video is not the way to teach your kid.

AL               Is that true of Sesame Street and other programmes directed at, say, three- to five-year-olds?

EO              Yes. Three- to five-year-olds can learn from TV and, actually, there’s some interesting research, much of which is by economists which is about how Sesame Street when it first came on increased school-readiness for some kids. So, it looks like there are things your kids can learn which of course is a caution because they can learn from other television shows, not Sesame Street also, so you’ve got to be a little careful of what they watch.

AL               My recollection is that the biggest effects, I’m not sure whether I’m thinking of Jesse’s paper or the Sesame Street studies, were among families from non-English speaking backgrounds, suggesting that some of the school-readiness is having English on the TV in a household where, perhaps, Spanish is spoken.

EO              Yes. I think that that’s right. I think that comes out in both Jesse’s work but also in the Sesame Street stuff. That in general these things are more important for kids from more disadvantaged backgrounds where the parent help on school readiness is not going to be as significant.

AL               Staying with the controversial topics. Parents who work. Having both parents work and also the impacts of day-care. What did you find about that?

EO              So, I think the data here is very reassuring in the sense that we worry a lot about what is best for the kid and the answer is these are all fine for the kid. It’s fine for both parents to work. Kids who have both parents work don’t have any different outcomes on average than kids who have a parent who stays at home.

I think that that opens up the possibility that we can, say, ask parents what they want and think about what is going to work for them. Which is something I don’t think we do enough.

Particularly with moms in the US, there is almost never an emphasis on what do you want to do and so one of the things I say in the book is, you should ask yourself what do you want to do? Do you want to work or not? When I’ve talked to women about that they’ve said, it’s amazing, no one ever asked me that. I went back to work. This is something you’re going to be doing every day for your whole life and you never thought about whether you would like to do it or not.

I think this is just illustrative of some of the issues that we have in this early parenting stuff, that people are just not thinking about what works for the family overall. They’re solely focused on the baby, which of course is in some ways the most important but also something where we do need to… There is a whole family, there is not just one person.

AL               And that’s again, I guess, coming back to your economist training, the notion of trade-offs of whether or not the effects for the child are small negative or small positive. You’d want to weigh that against an effect for the rest of the family which might be large positive.

EO              Exactly.

AL               I understand you didn’t want to write the discipline chapter at first and it was your editor who talked you into it. Tell us a little bit about what you ended up discussing, particularly that notion of discipline as education.

EO              I had just thought that there would not be any good evidence on discipline but she said, please look into this I’m sure that maybe we can find something. And it turned out that actually there is some good evidence on what kinds of discipline is most effective, particularly around the idea that discipline should be about trying to convey to your kid what is the right behaviour trying to educate them about what is the right behaviour.

                   There’s a lot of stuff on the importance of consistency, and saying following through on your threats, which should be, to be clear, not physical threats but removal of privileges. So, if you say, if you keep hitting your sister, you’re going to get a time out then you do have to give them a time out.

                   There’s some evidence that that kind of approach really does work. Some kind of warning approach. That’s a warning, that’s a second warning, that’s a time out. There’s good evidence behind those approaches.

                   What there is not good evidence for is spanking. Discipline as it is practised in an evidence-based way should not include physical punishment which has been shown to probably have counterproductive effects but certainly not to have good effects on kids.

AL               One of the other interesting lessons of it for parents who like to overthink is the notion that children aren’t adults, and sometimes overexplaining what’s going on can be counterproductive, which was insightful for me.

EO              Yes. Me too. I think I do this all the time that I want… I’m a person who likes explanations and so I have this instinct with my kids like, okay let me explain to you why that behaviour was not right. And, why don’t you want to keep your shirt on? Because adults keep their shirt on. I think a lot these say, hey, you’ve got to recognise that even your older kid, even an eight-year-old, a nine-year-old, is not an adult, and expecting them to react like an adult is counterproductive.

                   Some of this discipline stuff is just taking a step back for yourself and being like, okay, what can I reasonably expect from my kid here.

AL               Thinking about kids, overall, economists have this model of humans as looking to maximise their overall happiness. Yet we have pretty solid evidence that parents are less happy than non-parents, and that if you follow the same people over time that their point in time happiness while they’ve got kids is significantly lower.

                   How do we explain the paradox of people having kids? Are we, in fact, not trying to maximise happiness? Is there something wrong with how we’re measuring happiness in these studies? What do we learn from the research and what does that tell us about parenting more broadly?

EO              It definitely is the case that in these happiness studies, people with kids are less happy. You can see data… I talk some of my data on marital satisfaction which declines in the early years of parenting. There’s some good work on this. There’s a book called All Joy No Fun which is a nice book about the idea that kids are not fun. But I think there’s a thing that all of that misses. Which is, of course, if you ask somebody who is a parent what is the best thing about your life, they will probably tell you it is their kids, even though in the moment it is sometimes the more frustrating thing.

                   I think there’s a sense in which studying happiness like this, in this moment-to-moment way does miss something about the broader picture of what you’re getting out of your kids.

                   I think the other thing it sometimes misses, is I find with my kids, the highs with the kids, the good moments, are so good. It isn’t that they happen all the time, but the moments when they do happen are so much better than any other thing in my life. That I think in some ways, even a happiness survey that was pinging you very frequently might miss those moments. But those moments are so amazing. I think there’s something that we’re missing about the measurement of happiness in the experience of parenting.

                   But I also think that parents put a lot of pressure on themselves, and it probably does lower the amount of fun that you’re having. And I guess in some ways one of the hopes of the book would be to take some of this pressure out and take some of the feeling like, I have to be constantly performing as a parent. And parenting should just be about trying to achieve the maximum outcome for my kid, as opposed to some of parenting should be enjoying my time with my kids.

AL               Right. I think about the evidence around natural growth parenting compared to helicopter parenting. Helicopter parenting seems to produce better outcomes for kids, but perhaps natural growth parenting produces better outcomes for parents.

EO              Yes. And I’m not sure helicopter parenting is… I think the jury’s out a little bit on whether the helicopters and the snowploughs and all the different kind of vehicle parenting we could be doing is better.

AL               So, I want to wrap up, Emily, by asking you a few questions about yourself. What advice would you give to your teenage self?

EO              Oh, god. How much time do we have? I think, probably, just that it will get better. I don’t think I was a very happy teenager and I actually struggle with how I’m going to parent in that age range because I didn’t find it to be a great personal experience. I think when I look back on it I wish I had known, okay, you’re going to go to college and get a job and there are going to be people who you like who are going to be your people, you just have to wait.

AL               What’s something you used to believe, but no longer do?

EO              I think I used to… Before I wrote this book, I used to believe that data would always answer my questions and I think that this experience has been humbling in the sense of telling me, you know what, you don’t have all the answers and your approaches are not always going to be right. And I think it’s made me a little bit less bossy, at least around these issues. I’m still bossy about everything else.

AL               So you’re a little more humble in that… When it comes to that.

EO              A little more humble when it comes to data.

AL               You’ve also done something quite unusual for an economist in writing a paper which effectively debunked an earlier piece of your own research. Around the impact of Hepatitis B on what Amartya Sen referred to as the missing women in developing nations. Can you tell us a little bit about that pair of papers, how they came about and what it was like to do that backflip, effectively?

EO              Sure. So, when I was in graduate school, I wrote this paper which got a lot of attention. Which was about the idea that maybe some of the imbalance in gender across populations was in fact about Hepatitis B and differences in genders at birth associated with that disease. Rather than, say, infanticide or sex-selective abortion which was the prevailing view.

                   The paper got a lot of attention. It got a lot of criticism. It ultimately… Some evidence came out from some other people that suggested it might not be right. I then went and collected some more evidence, I think, to be fair, hoping that I would turn out to be right. But it seemed like I was not. So, I published a second paper which basically said, you know what, this first paper was not right. I am proud that I did that, I guess. I think that that was the right thing to do at the time.


                   It was a very difficult experience. Nobody likes to be wrong, and I was wrong in this very public way. I think that when I think back on the first paper, from the standpoint of someone who’s now been doing this for a long time, I definitely should have done more of the work that I did later in service of the first paper. I think that was a good lesson. Perhaps not an easy one. But ultimately that episode, like many episodes in our lives which are… We look back on as difficult ones, it was an important growing experience.

AL               Just to provide the listener with a little bit more context, your first Journal of Political Economy paper looked at the correlation at a national level between Hepatitis B rates and sex ratios. Your Economics Letters follow-up involved a study in which you actually had the Hepatitis B… Whether or not mum and dad had Hepatitis B, and then you looked at the gender of the children, so you were able to go straight down to the individual level using that Chinese data.

                   Did you have misgivings at the time about doing that study? And what would your recommendation be for other economists whose work is being challenged and who have the inclination, that I think most scholars have when their work is challenged, to hunker down and defend the results?

EO              Yes, I think that it was… I did have misgivings about doing it. For one thing, it took a lot of time and it was very expensive. Partly, I thought that maybe I would turn out to be right, but I understood that there was a chance I would turn out to be wrong and felt like I have to close this loop. We weren’t going to make… Ultimately, I felt that it needed to end in one way or the other.

                   I think there is an instinct to defend your work to the last drop, and that is a very human instinct, but I think we have seen some of this in economics where there’s been… These conflicts have drawn out in a way that is not productive and not focused and I think we might be better off as a profession if we tried to resolve those a bit faster.

AL               Emily, what’s the most important thing you do in your life to stay mentally and physically healthy?

EO              Probably Pilates. It sounds ridiculous, but I have an hour a week in which I go and do Pilates and I’m very strict about keeping that time, and I think that it’s not very much time. But it is important to me that I have some time in which I say, okay this is something I’m just doing for me. I’m not doing it for my job or for my family it’s just for me. I think that that’s very mental health… Valuable for my personal setting.

AL               Do you have any guilty pleasures?

EO              Coffee in the morning is about as guilty as it gets at this point.

AL               And finally, which person or experience has most shaped your view of living an ethical life?

EO              I think my husband, actually. My husband is a really ethical and principled person, sometimes to the point of tremendous frustration for me. But I think when you live with somebody for this long, it’s hard not to have some of them rub off on you, or at least to get a sense of the value of their values.

                   And he is really honest, and really does not back down when he believes in something and I think that’s something I’ve tried to internalise.

AL               That’s beautiful. Emily Oster’s two books, Expecting Better and Cribsheet, I cannot recommend them highly enough. A terrific gift for anyone you know who is expecting or has just given birth.

EO              Thank you, and Andrew, I…

AL               Emily, thanks so much for taking the time today…

EO              I should say that the new book is out in Australia next week, officially.

AL               Fantastic. Which may well be last week, by the time this podcast goes to air.

EO              Okay.

AL               But it will be in good bookstores and good book websites everywhere. Thanks again, Emily.

EO              Thank you Andrew.

AL               Thanks for listening to this week’s episode of The Good Life. We love getting feedback, so please leave us a rating or a comment on Apple Podcasts, formally known as iTunes.

                   Next week, I’ll be back with another inspiring guest, to discuss living a happier, healthier and more ethical life.


Be the first to comment

Please check your e-mail for a link to activate your account.

Stay in touch

Subscribe to our monthly newsletter


Cnr Gungahlin Pl and Efkarpidis Street, Gungahlin ACT 2912 | 02 6247 4396 | [email protected] | Authorised by A. Leigh MP, Australian Labor Party (ACT Branch), Canberra.