Smartphones and Birth Rates: What Teen Data Actually Shows

What Gets Conflated When Explaining the Decline in Birth Rates

South Korea recorded about 230,000 births in 2023, with a total fertility rate of 0.72.¹ When people try to explain the country’s falling birth rate, they tend to look at housing prices, childcare costs, jobs, and attitudes toward marriage and family—all together—because no single factor can account for the shift on its own.

Recently, some researchers have started adding smartphones and social media to that list. The hypothesis is that technologies which reshape how and how much time people spend meeting each other could also affect relationship formation and, in turn, childbearing.

But there’s a distinction worth keeping in mind when reading this kind of research. A decline in teenage pregnancy and childbirth is a fundamentally different phenomenon—different subjects, different meaning—from adults being unable to have children when they actually want to. The smartphone study I’m about to introduce mostly deals with the former.

Splitting First Births from Additional Births

A falling fertility rate can reflect two different shifts happening at once: fewer people having a first child, and people who already have children having fewer additional ones. Looking at the total fertility rate alone makes it hard to tell these two apart.

In a 2025 paper in Scientific Reports, Steven J. Shaw analyzed data from Italy, Japan, the UK, and the US, then extended the analysis to other high-income countries. He proposed a method for separately calculating, based on current age-specific birth patterns, what share of women would become mothers, and how many children each mother would have. This is a different metric from actually tracking real women over their lifetimes to confirm completed family size.

In this analysis, the US shows an indicator corresponding to first births declining from around 2008, while the indicator for children per mother stayed relatively stable or even rose. The author argues this means we need to examine the process of becoming a parent separately from family size itself.

But the mere fact that first births have declined doesn’t let us conclude that failure to form couples is the cause. Meeting a partner, cohabiting or marrying, timing of childbirth, and whether people want children at all are related to one another, yet each is a distinct stage that needs its own verification.

Conditions for forming a family can also vary by income, education, and employment status. Rather than applying the results from one particular group to every country and every woman, it’s more accurate to break down where, exactly, which change is occurring.

The draft looks accurate and complete. No Hangul, numbers preserved, structure matches. No edits needed.

A Study Comparing Mobile Network Coverage and Teen Births

There’s research that directly examines the hypothesis that smartphones change how people connect with each other.

Nathan Hudson and Hernán Moscoso Boedo of the University of Cincinnati analyzed U.S. network coverage and teen birth rates in a working paper written in April 2026. They also examined teen pregnancy indicators (under 18) in England and Wales. The authors present findings that the spread of high-speed networks accelerated the decline in teen births.

The researchers used regional variation in terrain elevation as an instrumental variable² affecting network coverage. This goes beyond simply comparing trends in smartphone penetration and birth rates — it’s a design meant to estimate cause and effect. That said, it requires assumptions such as terrain not affecting births through channels other than network access. And getting similar results in two countries doesn’t rule out the influence of other policy and economic conditions.

The paper also offers a cross-country comparison.

Using data from 128 countries, the researchers examined how age-specific birth rates changed before and after the spread of digital communication environments. For this, they used benchmarks like the timing of the iPhone’s launch and mobile subscriptions per capita. This isn’t data that fully measures individuals’ actual smartphone usage time.

In the comparison, the group showing the most pronounced change was ages 15-19. The effect was smaller for ages 20-24, and when accounting for each country’s overall birth trends, no clear change common across countries appeared for the 25-and-older group. This is why “Teen” in the paper’s title matters.

This age difference is central to the study, but it can’t be read simply as “younger people used smartphones more, so births declined proportionally.” Differences in daily life and relationship formation across age groups also need to be considered.

In Korea, the rapid spread of smartphones and the decline in the birth rate can be observed together. But two trends alone don’t prove causation. In particular, since births to women in their 30s make up a large share of Korea’s total, there are limits to explaining Korea’s overall low birth rate using results from teenagers abroad.

The Explanation: Smartphones Changed How Teens Spend Time Together

The pathway the authors propose is a decline in face-to-face interaction. Rather than dismissing existing explanations like access to contraception, they argue that the digital environment added a further push to a decline in teen births that was already underway.

When the peers you’d normally spend time with move online, your own use of time follows suit.

In the U.S. time-use data the paper analyzes, teens’ daily in-person socializing fell from 68 minutes in 2003 to 38 minutes in 2019. Over the same period, computer-based leisure rose from 22 minutes to 96 minutes. These figures are averages across the surveyed population.

This finding connects to the idea that fewer opportunities to meet peers in person may have reduced both sexual contact and unintended teen pregnancies. Extending the same pathway to explain adults’ declining rates of finding partners or having wanted children would require further research.

It’s also possible that expectations about relationships are shaped by online content. Still, an analysis of broadband access and teen births doesn’t go so far as to prove the claim that Instagram raised people’s standards for partners and thereby reduced marriage rates.

What matters isn’t just an individual’s own usage habits, but also where the people around them interact. If your friends mostly make plans and chat over messaging apps, you need to be there too if you want to keep up the relationship.

The authors’ model captures this kind of interaction. As more people use smartphones, online interaction becomes more convenient, which in turn can draw in still more users. Even someone who prefers face-to-face activities finds it hard to keep a different lifestyle when everyone around them has gathered online.

What TV Research Tells Us About Media’s Influence

The idea that media can shape how people think about family and childbearing isn’t new to the smartphone era.

Robert Hornik and Emile McAnany examined the relationship between mass media and fertility change in a 2001 study. They explain that it’s not just time spent with media, but the information conveyed, discussion among people, and shifts in values that work together. It’s not an argument that TV penetration alone explains all fertility change.

A 2012 study by Eliana La Ferrara and colleagues compared the timing of when Brazil’s Globo network rolled out across different regions. Globo’s telenovelas³ frequently featured families with few children, and the study found that fertility rates dropped further among women in regions that gained access to the broadcasts. The effect was larger among women of lower socioeconomic status and those in the later stages of their childbearing years.

This research shows that repeated exposure to a certain image of family life can influence people’s choices. Still, we can’t simply transplant the magnitude of these TV effects onto smartphones, nor can we assume that because smartphones are used for longer hours, their influence must necessarily be greater.

Alice Evans, who studies gender norms and social change, uses the phrase “cultural leapfrogging.”⁴ Her explanation: by encountering how people live in other societies online, individuals can start to reexamine expectations and norms they’d long taken for granted in their own communities.

This line of research suggests that digital technology has the potential to reshape how people think about family. But it’s not grounds for attributing fertility decline in any particular region solely to shifting expectations among women. We also need to look at the economic and social conditions required to form relationships and build families in the first place.

Oswarld’s Lens

Reading these studies, I was reminded of the service-adoption dynamics I used to study when building GTM strategies.

When more people use a service, that gives the people around them a reason to use it too. Think of how awkward it is to be left out of a group chat or a meetup because you don’t use KakaoTalk, a Korean messaging app. A product stops being a matter of individual choice and starts shaping how people coordinate and interact with each other.

Looking at it this way, what stands out to me is that the smartphone isn’t just a tool that changes how we spend time alone. It can also affect where people choose to meet each other. The teen studies give us grounds to explore that possibility, but they don’t answer every question about adult marriage and childbirth.

That’s why I don’t want to treat housing prices, childcare costs, values, and the digital environment as competing single causes. I think we need to look at how these factors work together. The way social media shows us other people’s lives could also play a role, but its size and direction need to be checked separately.

The conclusion that restricting smartphones would raise the birth rate doesn’t follow directly from these studies either. What I care about is creating the time and space for people to meet in person and sustain relationships when they want to. There are structural conditions that simply telling individuals to put down their phones can’t fix.

Closing

To really understand fertility rates, we need to separate first births from subsequent births, and look at how patterns shift across age groups. Just because one case shows a stable number of children per mother doesn’t mean every country’s low birth rate has the same underlying cause.

The smartphone research suggests that changes in how people communicate may have affected teenagers’ face-to-face interactions, and by extension, pregnancy and childbirth. But explaining low fertility among adults would require additional evidence beyond that scope.

Technology has multiplied the ways we can connect with each other — it’s worth also looking at how opportunities for meeting in person have changed alongside that.

Which relationships have gotten easier to maintain online, and which have become harder to sustain face-to-face, in your own life? That question might be a good place to start thinking about the difference.

References & Further Reading

Primary sources

• Nathan Hudson & Hernan Moscoso Boedo, “The Collapse of Teen Fertility in the Digital Era”, University of Cincinnati Working Paper, April 2026. : This is the core paper behind today’s newsletter — a working paper that mainly analyzes teen childbirth and pregnancy alongside telecom infrastructure rollout.
• Stephen J. Shaw, “On a microdemographic framework for decomposing contemporary fertility dynamics”, Scientific Reports 15, Article 30726, August 2025. : This paper proposes a method for breaking down current fertility patterns into age at first birth and number of children per mother.
• Financial Times, “Why birth rates are falling everywhere all at once”, May 2026. ; The FT’s own data analysis visually shows how the timing of smartphone adoption across countries lines up with inflection points in fertility rates.

Background

• 2023 Birth Statistics, Statistics Korea, August 28, 2024.

• Alice Evans, Cultural Leapfrogging: Swiping Past Tradition : An essay on the relationship between the culture and social norms people encounter online.

• Eliana La Ferrara, Alberto Chong & Suzanne Duryea, “Soap Operas and Fertility: Evidence from Brazil”, American Economic Journal: Applied Economics 4(4), 2012. : A study analyzing the regional rollout of Brazilian soap operas and its relationship to fertility.

• Robert Hornik & Emile McAnany, “Theories and Evidence: Mass Media Effects and Fertility Change”, Communication Theory 11(4), 2001. : A review of the various pathways and evidence through which mass media can influence fertility.

The author is Oswarld (Kwangseob Ahn). Current roles: Adjunct Professor at Sejong University, Strategy Consultant at INLEVEL9. Career, research, books, and recent work are kept current on the About page. Latest · July 2026: HEMA-2: A Consolidation-Aware Tri-Memory Architecture with Multi-Channel Scheduling for Lifelong Conversational AI.

Footnotes

1. Total fertility rate (TFR): The average number of children a woman would be expected to have over her lifetime if the age-specific birth rates of a given year held constant throughout her childbearing years. This differs from the actual number of children a generation ends up having. ↩

2. Instrumental variable: An analytical method used to estimate cause and effect. Here, it uses terrain differences that affect telecom network rollout. This requires assumptions such as terrain not affecting fertility through any other pathway. ↩

3. Telenovela: A Latin American TV soap opera. ↩

4. Cultural leapfrogging: A term Alice Evans uses to describe the phenomenon of people reconsidering their own society’s existing norms after encountering the culture and values of other societies online. ↩