| Aug 15, 2026
Falling birth rates are usually blamed on the cost of children. I think we have also made it too easy to spend our free time alone.
Housing, childcare, careers, and student debt are all real. They mostly explain why couples have fewer kids. They do less to explain why fewer people are ending up in couples in the first place. In a lot of rich countries, dating and marriage have been falling alongside fertility. That is a different problem, and the usual policy toolkit of tax credits and parental leave does not really touch it.
The idea I keep coming back to is this: modern life has been quietly privatizing leisure. Fun used to be something you did with other people, because that was often the only way to have it. Now some of the best entertainment in history fits in your pocket and asks nothing of you. Once leisure is cheaper, easier, and more reliable alone than together, you get fewer incidental conversations, fewer friendships, fewer relationships, and eventually fewer children. Nobody has to decide against a family. Thousands of slightly easier nights in can do it for you.
The relative price of a night in
Social leisure has always had transaction costs. You have to coordinate schedules, travel somewhere, spend money, and take on the small social risk that the night will be awkward or that someone will say no. Solo leisure increasingly has almost none of that.
The older version of "I want to have fun tonight" was: call friends, pick a place, get dressed, go out, interact with people. The modern version is: pick up the phone, open YouTube or Reddit or Netflix or a game, and get something immediately rewarding. The second option is cheaper, easier, less risky, and, year by year, better at being entertaining.
This does not require anyone to consciously reject relationships. It only requires that, on the margin, staying in keeps winning.
There is a second path that matters just as much. Individualized leisure also pulls people out of family and community life. You see fewer married couples, fewer kids, fewer extended-family gatherings. Parenthood becomes less salient. "I don't really want children" can be a genuine preference, or it can be what it sounds like when parenthood never became a normal part of the life you actually live. Those are not the same thing, but they produce the same fertility statistic.
This is not a smartphone story
The phone is only the latest and most powerful version of a much older trend.
In preindustrial life, leisure was mostly communal: festivals, religious gatherings, taverns, storytelling, neighborhood games, extended family. Books, newspapers, and radio made solitary entertainment more attractive. Television made it passive and always available. Cars and suburbs made spontaneous contact harder. Cable, personal computers, and video games made it more private still. The internet made it unlimited. The smartphone made it portable. Algorithms, streaming, games, and now AI can tailor it to one person.
The through-line is not "phones made people infertile." It is that leisure has been moving, for a long time, from something we consume together to something we consume alone. Fertility started falling long before smartphones, which is a problem for the cartoon version of this argument and not much of a problem for the actual one. The phone is an accelerant of a centuries-long shift: private alternatives to social life got better, cheaper, and closer at hand.
Urbanization is easy to misread here. Density by itself does not kill social life. Traditional dense cities and villages can be extremely social. What seems to matter more is the privatization of everyday errands. Walking to a market used to mean neighbors, street vendors, acquaintances, and conversations. Driving to a supermarket can mean house, private car, parking lot, store, private car, house. The same economic activity produces wildly different amounts of incidental human contact. Delivery, remote work, and online shopping push that even further. Modernization keeps deleting situations in which you are forced to deal with other humans.
Two kinds of low fertility
It is worth separating two mechanisms that look identical in a birth-rate chart.
One is preference: people form relationships and then choose few or no children. The other is a relationship-market failure: people never form the relationships in which children would be born, so parenthood never becomes a real choice. A person who says "I don't want children" may mean that. Another person may mean "I don't see children fitting into the life I have." Policy that only cheapens childcare addresses the first group.
Preferences themselves may be downstream of the social environment. Imagine two people. One grew up with siblings, cousins, a religious community, married friends, neighborhood kids, and frequent family gatherings. Parenthood is baked into their picture of adulthood. The other grew up with a small household, solitary hobbies, online friends, little contact with children, mostly unmarried peers, and a lot of individualized entertainment. Parenthood may simply never feel like it belongs to them. The second person can be perfectly sincere about not wanting kids without that preference being a deep, fixed fact about their personality.
Once fewer people have children, the environment feeds on itself. Fewer kids in everyday life means fewer people interact with children, which makes parenthood less familiar, which means fewer people envision themselves as parents. The same kind of loop can run through relationships: fewer couples means fewer social events involving couples, which means fewer chances to meet someone. And through leisure: better solo entertainment means more people remain single, which makes being alone feel more normal, which justifies still more investment in solo entertainment. Small initial shifts can get large.
Phones in poor countries are not the same phones
A fair objection: plenty of poorer countries now have high smartphone penetration and still have relatively high fertility. Owning a phone is not the same thing as living inside a developed, private entertainment ecosystem.
Data can be expensive relative to income. Fixed broadband may be rare. Electricity may be unreliable. Phones may be shared, low-end, or used mostly for work, payments, messaging, and navigation. Streaming video is data-hungry. Private space for solitary leisure may be scarce, while extended family and dense community still supply a lot of low-cost social life. The useful question is not "does this country have smartphones?" It is how much high-quality, private, low-marginal-cost solo leisure the average person actually has.
Israel, India, and South Korea
Some places that look like counterexamples may actually be evidence.
Israel has unusually high fertility for a rich country. The usual explanation is strong family orientation, which is true and not a competing story. The more interesting question is what keeps that orientation in place. For observant Jews, Shabbat is a weekly, institutional interruption of the normal technological environment: work stops, commerce stops, electronics are restricted, family meals become central, and community participation rises. Instead of each person independently deciding whether to socialize, the institution creates a recurring stretch of time in which people are the entertainment. Family, community, marriage, children, and religion reinforce each other every week. Israel may not be a country that escaped this hypothesis. It may be a country with an institution that keeps rebuilding the social environment the hypothesis says we are losing.
India is a different kind of qualification. If fertility falls partly because incidental social interaction dries up, arranged marriage can bypass the bottleneck. Western partner-finding looks like: social exposure, meet someone, date, marry, children. An arranged-marriage system can look like: family networks identify candidates, families evaluate compatibility, marriage, children. Declining everyday socialization should matter less when a society has an institution whose job is to produce matches. The prediction is uncomfortable and testable: as India gets richer and more individualized, fertility should fall especially hard if arranged marriage weakens and partner-finding shifts onto individual dating.
South Korea is the case where several of these forces seem to stack. High technology, abundant individualized entertainment, collapsing fertility, falling marriage, and less dating among young adults. A 2024 paper by Lee, Brzozowska, and Sobotka, using nationally representative Korean surveys, reported that in 2018 only about 26% of unmarried men and 34% of unmarried women aged 20–29 had a dating partner — roughly 30 percentage points lower than in 1991. Korea also moved from stronger family-mediated matchmaking toward individual "love marriage" at the same time that solo leisure was getting better and incidental interaction was getting rarer. That is a nasty interaction: the society asked young people to find their own partners just as everyday life was giving them fewer chances to do so.
Gender polarization may be a second, independent force. Changing expectations, economic pressure, and cultural conflict can make men and women less willing to deal with each other. If people are both spending more time alone and becoming less willing to interact across gender, the relationship market can deteriorate very fast. I do not think Korean culture simply prevents men and women from mixing — many Koreans have opposite-sex friends. The sharper fact is that the pool of people actively forming those relationships appears to be shrinking.
The strongest version of the pattern, then, is not technology by itself. It is high individualized leisure plus weak social institutions plus partner-finding that depends on individual search. Korea looks like that overlap. India still has the matchmaking institution. Israel still has the weekly social one.
What would have to be true
The central empirical question is not "do people want children?" It is how much of the fertility decline comes from couples choosing fewer children versus fewer people forming the relationships in which children would be born. If the second piece is growing in wealthy countries, this theory is doing real work.
A few predictions follow. The link between individualized leisure and fertility should be stronger where marriage depends on individual partner search, and weaker where institutional matchmaking is still doing that job. Institutions that force recurring social contact — religious communities, communal meals, sports leagues, military service, residential universities, extended family, neighborhood groups — should partly protect fertility. High smartphone ownership in poor countries should not predict low fertility unless it also comes with the broader private-entertainment stack. And exposure to children and family-oriented networks should move stated fertility preferences, not just behavior.
If any of that is right, conventional fertility policy is aimed at the wrong bottleneck. Childcare subsidies, tax credits, parental leave, and housing help with the cost of having children. They do not answer why fewer people are entering relationships in the first place. The more useful direction might be rebuilding the boring social infrastructure that used to make friendship and romance a byproduct of ordinary life: clubs, third places, walkable neighborhoods, communal events, less geographic isolation. The goal would not be to talk people into having kids. It would be to make an environment where friendship, romance, partnership, and family can happen without anyone having to schedule them as a project.
The thesis without phones
The modern fertility decline may be partly a consequence of the progressive privatization of leisure and social life. As technology, wealth, urban form, and individualization make solitary activities more attractive and social interaction more costly, people get fewer incidental interactions and fewer chances to form relationships. At the same time, weaker participation in family and community life may thin out the social formation of family-oriented preferences themselves. The result is not simply that people sit down and choose fewer children. Increasingly, fewer people enter the relationships and social worlds in which having children becomes a natural possibility.
The smartphone is just an especially good recent example of that process. The fertility crisis may be, in part, a social-interaction crisis.
The hypothesis is mine. I worked it out in conversation, then asked an AI to synthesize my notes into this essay.