This series is adapted, with permission, from a Chinese-language series by our friend Marcus Ji, who looks at historical events through an economic lens. We have condensed and adapted the original for an international audience.
Fewer people are getting married, and fewer people are having children. The two are usually explained separately, and each explanation has its own long list of culprits: house prices, women’s independence, individualism, the education arms race, the cost of weddings, punishing work hours. Every item sounds plausible; stacked together they feel like a pile of fragments, with no one able to say which is the real cause.
Underneath the fragments, though, runs a single, simpler economic throughline that explains both at once – and also explains a fact that baffles governments everywhere: why paying people to marry and have children almost never works.
The throughline is this. For most of history, marriage and children were both economically worthwhile – not romantic choices but arrangements that paid for themselves. The modern economy has, in a series of specific and nameable ways, stripped away that payback, one piece at a time, until both became consumption – things you have because you want them (love, company, the joys of family), not because you need them. And once something turns from “must have, and pays you back” into “buy it if you feel like it, for the feeling,” it becomes optional: it can be postponed, discounted, or skipped. The fall in marriage and birth rates is, at bottom, the result of that one shift.
Marriage was once a small firm
Today we assume marriage is for love. But for most of human history it was first of all an economic arrangement. The economist Gary Becker compared the traditional family to a small firm, whose competitive edge came from specialisation – one person concentrating on earning, the other on the home and children, each becoming efficient at their task, so that the two together produced more than each would alone. That was the first source of marriage’s payback.
The second was insurance. In an age with no social security, no pensions and little access to credit, marriage was an ordinary person’s most important protection against disaster: if one spouse fell ill or lost work, the other carried the household; two families joined by marriage spread risk across a larger pool. Economists call this the risk-sharing function – a policy the poor could afford.
The modern economy dismantled both:
- The gains from specialisation evaporated as women could earn. Once women were educated, employed and able to earn for themselves, the “you earn, I keep house” split stopped being worthwhile – two earners extract far less extra benefit from dividing labour that way. As the economists Betsey Stevenson and Justin Wolfers put it in 2007, the marriage built on one person specialising in the home and the other in the market is dying out as women’s education catches and overtakes men’s.
- The insurance function was replaced by welfare, credit and women’s own income. With pensions, health cover, consumer credit and a salary of one’s own, a single person can absorb risk unaided. Marriage’s value as an insurance policy fell accordingly.
What is left once both are stripped away? Stevenson and Wolfers gave it a name: the marriage of shared consumption – people marry now for the film watched together, the children raised together, the interests and the life shared. That is love and companionship, and it is a genuinely good thing – but it has two awkward features. It is optional (you can live without it), and it favours those with the time and money to invest in a relationship. Marriage has quietly become something closer to a luxury.
Children were once a worthwhile investment
The same story plays out with children, more starkly. As early as 1929 the philosopher Bertrand Russell noted the economic chain: before factory laws banned child labour, a child could work and was a source of household income; once the laws passed, the child turned from an earner into an expense.
The demographer John Caldwell built this into a theory – wealth flows – whose insight is that what decides how many children people want is the direction wealth flows between the generations.
In a traditional farming society, children work from a young age and support their parents in old age, so wealth flows upward, from children to parents. An extra child is an extra worker and an extra pension – childbearing is a sound investment, and the rational choice is to have many.
In a modern society, children no longer earn, old age is covered by pensions, and the education arms race turns each child into a vast outlay, so wealth flows downward, from parents to children. The child has gone from asset to pure cost – and the rational choice becomes fewer children, raised more intensively.
Three developments stripped away children’s payback at once:
- child-labour laws removed their labour value;
- pensions removed their old-age value;
- universal education turned “have a few more” into “have fewer, invest more” – what economists call the quantity–quality trade-off.
One honest caveat: Caldwell’s strong claim – that children were a net financial gain in traditional societies – has not held up well to later evidence; some studies find that even in high-fertility societies a child’s labour may not fully repay the cost of raising them. But the robust half of the theory stands firm: as raising children gets more expensive, people want fewer of them.
One transition, two declines – and cash can’t fix it
Put the two sections together and the throughline is clear: marriage and children went through the same economic transition – from a necessity that paid for itself to a consumption good you buy for the feeling. Because it is one transition, marriage and birth rates fall together – not two independent problems, but two faces of one.
This also explains the fact that most puzzles governments: why does paying people to marry and reproduce barely work? Because a subsidy changes only the price, while the transition removed the function. You can hand out a large cash bonus for a child, but you cannot refund the vanished economic functions – “children will support me in old age,” “marriage is my only insurance.” South Korea spent roughly US$300 billion over two decades, and its fertility rate fell to the world’s lowest, about 0.72 – because money can shave a little off the price but cannot buy back a use that a thing has lost.
There is a subtler reason cash falls short. Marriage and children are time-intensive consumption – the largest cost of enjoying them is not money but time, especially women’s time. Economists call this the opportunity cost: the earnings you give up to do something. As women become more educated and better paid, their time grows more valuable, so the time-hungry business of raising children and tending a marriage becomes, relatively, ever more expensive. That is why this is unlike an ordinary luxury, where the rich buy more – here, the richer the society and the more precious its people’s time, the fewer children they tend to want.
Where the function survived, the decline is milder
A claim like this has to survive being turned around: if “investment becoming consumption” is really the cause, then where that transition hasn’t happened, marriage and childbearing should not have fallen. That is exactly what we see.
Where children’s economic function still exists, people still have many. In societies still built on subsistence farming, children remain real labour and a real basis for old age – wealth still flows upward – and fertility stays high. Because children still pay back, people still want them; without the step from asset to cost, the collapse in fertility simply does not occur.
And where culture steps in to replace the lost economic function, richer can mean more children, not fewer. The clearest case is Israel: GDP per capita above US$50,000, yet a fertility rate around 2.85 – the one true outlier among developed countries. The reason is not economic but cultural: an unusually strong family and religious culture that supplies a powerful emotional and meaning payback for having children, standing in for the economic payback that vanished long ago. When the emotional return on the “consumption” is pulled high enough, people keep buying even when it doesn’t add up economically. What decides whether people have children, in the end, is whether the thing still pays back – economically, or emotionally.
The conclusion: not that people stopped caring, but that they stopped needing
We are used to blaming the decline on a long list of villains – house prices, the arms race, feminism, individualism. But these are mostly the expressions and accelerators of the transition, not its root. The root is plainer: marriage and children were once economic necessities that let people survive and live better, and the modern economy has turned them into consumption bought for love and meaning.
Fewer marriages and fewer children does not mean people value them less – if anything the opposite, since as “consumption goods” people now expect more of a marriage’s quality and pour more into each child’s upbringing than ever before. What changed is not how much people cherish these things, but how much they need them: people no longer need marriage to pool risk, or children to support them in old age. And once something is downgraded from a need to a want, its quantity must fall – because a want can be postponed, discounted, or dropped.
Which is why any attempt to coax people back to marriage and childbearing with subsidies or policy is doomed to poor returns unless it restores some genuine payback – whether economic, or, as in Israel, cultural and emotional. You cannot use cash to buy back the fact that a thing is no longer needed.
Previously in this series: How did gender-based division of labour come about in the first place?
Next in this series: Is it true that successful women in Asia find it harder to marry?
