"We See a Decline in the Share of Women Wanting Children": Talking about Fertility, Family Choice, and Economics

Tom Heintjes: Hello, and welcome to another episode of the Economy Matters podcast. I'm Tom Heintjes, an editor in the Atlanta Fed's Public Affairs department. Today, we're sitting down with Julie Hotchkiss, a research economist and policy adviser here at the Atlanta Fed, and we're going to discuss her recent research into the complex—and, in some ways, even sensitive—area of birth rates. Julie, along with a number of fellow researchers, has written a paper titled "What's Behind Declining Birth Rates in the US?" So, without further ado, let's say hi to Julie Hotchkiss. Julie, thanks for being here today. It's great to have you back on the podcast—it's been a while.

Julie Hotchkiss: It has indeed! Thanks, Tom. I always enjoy talking about my latest research projects, especially with you.

Heintjes: And just so you know, I'm fighting the temptation to say this research is pregnant with possibilities for our discussion—well, fighting and losing, I guess.

Hotchkiss: Well, no worries. This topic does perhaps lend itself more easily to cocktail conversation than policymaking at the Federal Reserve Bank. But, as we'll get into, there are important economic implications.

Heintjes: Yes, there are, and they're really interesting—and I'm going to dive right in, since there's a lot for us to get to. First of all, your paper notes that birth rates have been falling worldwide for decades. Should we make a distinction between where they're falling, and where they're not—advanced economies versus emerging ones?

Hotchkiss: Well, the timing does differ across countries somewhat, but the decline truly is something that's shared around the world right now. There's a broad literature that explores macroeconomic factors related to birth rates overall, and the timing of their declines across different countries. But the focus of our research is more of an individual or microeconomic level, and it uses data only from the US.

Heintjes: I'm always curious about how research topics come to someone's attention. What put this one on your radar?

Hotchkiss: Like many of my projects, this one came out of a conversation over dinner. As baby boomer parents are wont to do these days, we were discussing with our friends when, or if, any of our millennial children would ever have children of their own. Now, of course, everyone's aware that younger generations of women are not having as many children as their parents and grandparents, and the general consensus, generally, is that declining birth rates are being driven completely by choice. But one of the people around the dinner table was a family formation lawyer, and her impression is that women are actually having more difficulty in getting pregnant these days, hence the need for her assisted-reproduction lawyer services. And then there's another dinner guest, an environmental lawyer, who followed up what she had to say and said that he thinks the increased difficulty is actually related to the environment, and specifically our increasing exposure to microplastics. And then naturally I said, "Well, let's find out." And these two lawyers, by the way, ended up being coauthors on our paper.

Heintjes: This might be the first time I've heard of someone seeking to involve lawyers.

Hotchkiss: Well, they didn't say anything about billable hours, so I figured we'd be safe.

Heintjes: So smart. Julie, you consider a number of metrics in this paper—age, ethnicity, and environmental factors, just to name a few. What data sources did you use for this paper?

Hotchkiss: Well, the primary data source that we use is called the National Survey of Family Growth [NSFG]. It's a survey conducted by the Centers for Disease Control and Prevention, and this survey is the most comprehensive, long-running survey in the US about pregnancy, births, marriage, cohabitation, contraception, fertility, and general reproductive health. The survey is run by the CDC's National Center for Health Statistics, and they've actually been collecting these data for over 50 years.

Heintjes: That is a lot of data. Can you briefly describe the survey?

Hotchkiss: Well, each survey that they do includes, I think, between 6,000 and 10,000 women between the ages of 15 and 49. We restrict our analysis to women between the ages of 20 and 40 years old, mostly so we can compare women at the same age across multiple generations, and that age range allowed us to see multiple generations. But most importantly, the survey captures three really important pieces of information for the analysis. The first one is related to women's behavior, and this is whether or not they've had any children. The second piece of information is related to their preferences—whether they want children—and then the third is their physical constraints—whether or not they're having any difficulty getting pregnant, or carrying a pregnancy to term.

Heintjes: That is an amazing repository of information, Julie. Apart from supplying data for you to use in your research here, what do people do with this huge trove of information?

Hotchkiss: Obviously, our research only scratches the surface of topics someone could research using these data. For example, someone might want to know how children's mental or physical health is impacted by family transitions, like a parental divorce or remarriage. Or they might be interested in the importance of access to insurance for addressing mental health issues, or what factors impact a family's spacing of having multiple children—or actually, we could probably spend the whole podcast just exploring all the other possible research questions these data can be used for.

Heintjes: Absolutely. As you noted, though, for the sake of our conversation today, we're going to confine ourselves to the US. In your paper, you referred to "replacement rates"—which economists pay a lot of attention to. And that's the rate required to keep a population constant, or increasing somewhat. Is that right? Does that mean births only, or does it also include migration from other countries?

Hotchkiss: Well, this is a good question. The replacement rate is really just a birth-based benchmark. In countries where infant mortality rates are low, like in the US, we estimate that the replacement rate is about 2.1 children per woman, on average, in order to keep the population stead, absent migration. So falling below that number of children slows population growth—unless, of course, immigration fills the gap.

Heintjes: And as your paper notes, this dwindling birth rate has major economic implications, doesn't it?

Hotchkiss: It sure does—and timing matters. In fact, in the short run, lower fertility—or women having fewer babies—can free up women to supply more labor to the economy, and this would boost economic growth in the short run. But in the long run, persistently low fertility reduces aggregate demand and innovation potential. This will weigh down economic growth, and as the population ages this will strain programs like Social Security.

Heintjes: Your research touches on the long history of differing perspectives on fertility rates and economic growth, but it wasn't always an unquestioned assumption that more people equals greater economic prosperity, was it? Some economists used to maintain that a larger population put a strain on finite resources, and so forth.

Hotchkiss: Yes, you're absolutely right. The view that growing populations will put increasing strain on resources goes all the way back to Malthus of the 18th century. He argued that populations grow at a faster rate than resources do, spelling doom for our civilization. In fact, it was Malthus's theories that are the reason that economics is often referred to as the "dismal" science. His theories, though, underestimate humans' abilities to innovate and adapt.

Heintjes: So is the debate about how fertility affects economic growth settled, then?

Hotchkiss: A debate among economists settled? That's so cute, Tom.

Heintjes: Touché.

Hotchkiss: I wouldn't necessarily call it settled, but the most recent compelling evidence points to what I referred to as a nonlinear relationship, with lower fertility boosting labor supply and economic growth in the short run, but having a longer-term—and longer-lasting—negative consequence for growth. And this is further evidence that economists really do have two hands. The answer just depends on your perspective.

Heintjes: Right. The long-run negative impact on economic growth is because with fewer babies, there will be fewer adults in the future demanding and consuming products?

Hotchkiss: Exactly.

Heintjes: So, how much has the birth rate in the US changed over time?

Hotchkiss: Well, we look at the United Nations for these data, and we see that total fertility—which is what we compare to the replacement rate—peaked in the US in the early 1960s, at about 3.5 children per woman. Then it fell steadily and hovered right around the replacement rate of about 2, 2.1, until the mid-2000s. Then it started falling again. Total fertility in the US right now sits at about 1.5 children per woman. The lowest fertility rate in the world is in South Korea—it's at about 0.65 children per woman.

Heintjes: Well, what's behind the declines in birth rates since the 1950s?

Hotchkiss: Well, there's two primary contributors credited with driving the decline in birth rates, starting in the 1960s. In 1960 specifically, this is the year that oral contraception was made available, giving women more agency about controlling their own fertility. And although we typically point to the 1973 Roe v. Wade Supreme Court decision as a turning point in women's access to abortion services, in reality, several states were already expanding access before the Supreme Court's decision.

Heintjes: I guess I'm going to ask you to generalize here: Do these declining birth rates cross demographic and socioeconomic lines?

Hotchkiss: Yes, they absolutely do. Others have shown that these declines reflect a widespread decline in fertility rates among women with different education, race, marital status, and so on.

Heintjes: Interesting. Julie, getting back to the data set that you use, the National Survey of Family Growth—are those data specific to certain age cohorts—Silent Generation, baby boomers, Gen X, and so forth? And what does that survey tell us about changing attitudes toward starting a family?

Hotchkiss: This is actually one nuance that we add to the use of the data. The CDC simply reports responses by individuals who completed the surveys, and then they also put out reports—but they just compare changes in responses over time. But women in 1982 are not the same women being surveyed in 2023. So most importantly, the age distribution, the education, the marital status of women over this time period changes a lot.

Heintjes: That's a great point. It's not much use to compare apples to oranges when you're discussing things like fertility decisions across age groups, right?

Hotchkiss: Exactly. Fertility decisions and the difficulties with pregnancies are very, very age specific. If we want to assess how women are changing over time in terms of these two mechanisms—the decisions that they make, and the difficulties they're having—we need to compare women over time who are at the same age. So, this is why we present our results by both cohort and age. I'll just give you a little bit more detail. For example, when we compare how many baby boomer women say they want children versus Gen Z women, we need to ensure we're making that comparison for women across these cohorts, but when these women were of the same age.

Heintjes: Not that you have to be married to have a baby, but it's established that marriage rates are also down among younger adults. Would I be off base to connect those dots?

Hotchkiss: Not at all—those dots are quite connectable. We also observe a declining share of women who are partnered or married in the NSFG data set, and it's true that women with a partner are more likely to have a child at any given point in time. But our results show that the declines in having children and wanting children, and increases in the difficulty having children, are pretty much the same. The patterns are the same among married women as they are among single women. So while declining marriage rates—you're right—are definitely contributing to declining birth rates, it's clearly not the whole story.

Heintjes: How has the rate of women wanting children changed over time? As daycare and all the costs of having a large family go up and up and up, are more women simply opting out?

Hotchkiss: We definitely do see not only the well-documented decline across the years in the share of women having children, at any given point in time and at any given age, but we also see a decline in the share of women wanting children. And at the same time, we're seeing a rise in medical difficulty bearing children, and this is the crux of the research question. When we compare two women at the same age, education, marital status, and so forth, how does wanting children stack up against any difficulty bearing children in explaining whether a woman ends up having any children at all? In other words, is the decline in having children explained by desire, or difficulty, or both?

Heintjes: Does your research show any difference between women who are employed and have invested a lot in their careers, and women who may not have gone that route?

Hotchkiss: Well, this is an interesting point, actually. Our results show that women who are employed, or have more education, are more likely to want children—but they're less likely to actually have children. And this is what we would expect, based on other research showing a significant what we call "motherhood penalty." What this is, is that working women who have children are shown to suffer—in terms of wages and career advancement—this penalty, relative to their childless counterparts, and relative to men who become fathers. So, the more education and investment in a career that a woman has, the greater is the opportunity cost—what they have to give up—to have children. In other words, there's actually more to lose from the motherhood penalty among more educated women who have invested in a career, than there is for less educated women.

Heintjes: Julie, your paper says—and I quote here: "At every age, newer generations of women are less likely to have any children than generations that came before." That's pretty unambiguous. What do you think is behind this shift in mindset?

Hotchkiss: The main conclusion from the analysis is that it's not just one thing, and it's different across cohorts of women.

Heintjes: Yes, it's rarely one thing, right?

Hotchkiss: Exactly—especially from economics research. Anyway, both changes in the desire to have children and in the ability to bear children are factoring into the outcome of fewer children.

Heintjes: What did you find about changes in the desire to have children?

Hotchkiss: This is actually one result I found particularly striking. If we're looking at the share of women who say they want children, just over time, we see that the share is declining. A decreasing share of women say they want children from one year to the next now, as compared to the past. And part of the reason for this, of course, is that women are getting older, and older women are less likely to want children than younger women. But our analysis shows that this decline is all concentrated among Gen Z. That's the youngest generation we have in the data. In fact, Gen X and millennials—the two generations before Gen Z—actually have higher desire to have children than boomers did at the same age. So, for those cohorts, the Gen X and millennials, the decline in births isn't about wanting kids. But for Gen Z, the picture changes dramatically.

Heintjes: Wow. So how do those changes in wanting children compare to the change in difficulties having children?

Hotchkiss: Well, here the result is more consistent across cohorts. We found that impaired fecundity is higher for every cohort of women at every age, since baby boomers. What this means is that women in recent cohorts are having more difficulty conceiving or carrying a pregnancy successfully to full term. So taken together, declining desire to have children contributes to the fall in birth rates among Gen Z only, but increasing difficulty having children contributes to the fall in birth rates observed among all recent generations.

Heintjes: And why do you think the desire for Gen Z women to have children has declined so dramatically?

Hotchkiss: Since the fall in desire to have children is concentrated among Gen Z, we need to ask ourselves, "What is it about the environment, the experiences, and constraints facing that generation that would produce such dramatic change in desire?" And first, I want to make clear that the discussion, at this point, that we offer in the paper is purely suggestive. It's not causal.

Heintjes: Duly noted.

Hotchkiss: But the first point that we make—and you touched actually on this in one of your earlier questions, Tom—is that the financial burden of raising children is increasingly significant. In fact, seven out of ten Americans today say it is unaffordable to have children. Actually, this is a 13-point jump in just one year. And on this point, we do offer actually some evidence from our data that financial considerations are more important for Gen Z than earlier generations. And to be fair, income and financial considerations are important across all generations, but more so for millennials and Gen X than boomers. But Gen Z women elevate the importance of finances to a whole new level, and this is likely because the costs you mentioned have risen so dramatically.

Heintjes: And the second point?

Hotchkiss: Right—the second point is that Gen Z is often referred to as the "lockdown generation," and this is due to their disproportional exposure to school shootings and related trauma. We speculate that this exposure may either directly or indirectly cause them to question whether they want to bring their own children into this environment. We also point out, actually, when we're talking about fears and costs, that maternal mortality in the United States is at 17 deaths per 100,000 live births, and this is a higher rate than in any other developed nation—and it's even higher than in some less developed nations. So this—maternal mortality fears, potentially—and fears from violence facing their children, on top of other financial considerations, may be pushing Gen Z women over the edge of that decision to not want children.

Heintjes: You mentioned something I'd like to touch on again—the increasing difficulty having children affecting all three recent generations. What do you think is causing that?

Hotchkiss: Well, again, in answering this question, I want to be clear that our analysis doesn't uncover any causal contributors to rising impaired fecundity or difficulty bearing children. However, we do present suggestive evidence that environmental concerns could be a contributing factor. There's strong evidence provided in the medical literature that endocrine-disrupting chemicals negatively impact not only female but also male fertility, and we point out—again, using evidence from other sources—that the timing of the rise of increased exposure to microplastics, which actually contain endocrine-disrupting chemicals, is notable when overlaid with the rising difficulties they're experiencing having children. For example, we show in the paper that concentrations of plastics in the oceans and other consumer environments began rising sharply in the early 2000s, which is precisely when millennials and Gen Z were moving into their key reproductive years. This suggests that the more recent generations have had growing exposure to chemicals that are known to disrupt fertility and pregnancies.

Heintjes: We've talked about a lot of findings so far in our conversation. Did any of them come as a surprise to you?

Hotchkiss: I think that the biggest surprise to me was finding that, once we controlled for other observed factors, the desire to have children among millennials and Gen X was actually slightly higher than it was for baby boomers—again, comparing women of the same age. I guess I've gotten so used to blaming millennials for everything wrong in the economy, I thought family formation decisions surely would just be one more thing to add to the list, but it wasn't the case here. But I also want to point out that, in spite of wanting to have more children as much as their parents, these generations of women are facing increased physical difficulties and financial constraints.

Heintjes: So our listeners should note that Julie's list of things to blame on millennials has actually gotten shorter. Julie, what takeaways would you like policymakers to come away with after reading your research?

Hotchkiss: Well, first of all, my children will be delighted for you to have pointed out that this is one thing I don't blame on millennials. But to your question, the first thing that policymakers need to do is decide whether declining birth rates is a concern they actually want to address. If they agree that declining birth rates could threaten population growth and hence long-term economic growth, then we present and discuss a whole host of policy options that could each address individually identified potential contributors to the declining birth rate. Our identification of potential policies is obviously far from exhaustive, but we do touch on policies from social support through family leave policies, to environmental regulation on the use and disposal of plastics—also to supporting assisted reproductive technologies, such as IVF, and providing for expanded reproductive control. There are a lot of candidates, and I think it's safe to say that no one single policy will do the trick.

Heintjes: This next question might be a bit of a digression, but I feel the need to ask it. What about immigration? Would allowing more immigration into the country address the population growth concern?

Hotchkiss: Well, actually, this is a great question, Tom. The Congressional Budget Office estimates that because of the dramatic decline in net immigration in the US in 2025, if those immigration levels stay the same, the US population will stop growing by 2056. But I also want to come back to a point you made earlier on in our conversation, and that was that declining birth rates is not isolated to the US. Pretty much all countries are facing a similar decline. So, it's not clear, if we want to reverse our immigration policy, whether it would be successful in potentially contributing to our population growth.

Heintjes: Am I putting you on the spot by asking what you think the ideal solution, or solutions, would be to reverse falling birth rates? If you could wave a magic wand, what would happen?

Hotchkiss: Well, as I mentioned earlier, there really is no single policy, or magic wand, that would reverse falling birth rates—at least, that we've identified. But to put this baby to bed...okay, maybe that's a bit much...

Heintjes: That's a bit on the nose.

Hotchkiss: But anyway, I think the most important message from our research is that, in addition to the usual suspects—and those would be the financial incentives, for example—policies addressing the increasing medical difficulties facing women who do want to have children needs to have a seat at the table when discussing what's going on with birth rates.

Heintjes: And Julie, that sounds like a great note to end our conversation on. Thanks so much for being on today and sharing your research insights with us. I told you this would be a fertile area for conversation.

Hotchkiss: Well, played, Tom. Thank you—and you're welcome. This really has been a blast.

Heintjes: And I should note that we'll have a link to Julie's paper on our website—our newly redesigned website, I should note—at atlantafed.org, and I encourage you to check it out. It's a timely and wide ranging—and thought-provoking—piece of research. That's all for this episode. I hope after you're done reading Julie's paper, you'll check out the rest of the Atlanta Fed's website, where we have lots of information about economics, the economy, banking and finance, payments, and lots more. Thanks again for spending time with us today, and please come back next month for another episode.