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.
There is a stubborn belief, especially strong across East Asia, that the more successful a woman becomes, the harder she finds it to marry – the phenomenon China labels “leftover women.” The claim is rarely actually calculated. This piece follows the economics of mate selection – turning each criterion into “how many people you have to look at to find one” – and arrives at a counterintuitive answer: yes, the pool really does shrink, but not for the reason everyone assumes, and it is not the dead-end it looks like.
How each standard shrinks the pool
Whether a candidate qualifies comes down to three roughly independent filters, multiplied together: are they in the acceptable age range, are they single and available, and do they clear your standard. Run the numbers – Marcus does it in detail on a Chinese salary baseline – and two things stand out.
First, on age and availability the sexes are nearly symmetric; both are limited by the same arithmetic. Second, the asymmetry appears only at the third filter, income, and it comes from a well-documented behavioural difference: women tend to apply an income floor – he should earn at least as much as me, ideally more – while men, in field experiments on real dating platforms, apply essentially none. (A stated income multiplies a man’s inbound interest severalfold; a woman’s income barely moves a man’s interest at all.)
That single filter is what collapses the pool. On the same baseline, a woman who asks only that a partner not lower her living standard finds roughly 1 in 16 men qualify; raise the bar to “clearly earns more” and it drops to about 1 in 38; raise it to “top 10% earner” and it falls to around 1 in 87. A man applying no income filter stays near 1 in 12 throughout. Nothing about age or biology creates the gap – it is created entirely by a filter women tend to impose and men mostly do not.
When the target is a millionaire
Push the standard to its extreme and the pool doesn’t merely shrink – it crosses into another order of magnitude. Take net worth instead of salary. By China’s 2024 Hurun Wealth Report, against roughly 494 million households nationwide:
- households with total assets of ¥10 million or more (about US$1.4 million): about 1.68 million, or 0.34% – 1 in 290;
- with ¥10 million in investable assets: about 891,000, or 0.18% – 1 in 550;
- with ¥100 million or more: about 110,000, or 0.022% – 1 in 4,490.
And those are households of all ages (the head is usually 40–60), with the wealth mostly built by the previous generation. Add the filters a marriage-minded search actually needs – young (30–40), single, and in control of the money themselves rather than their parents – and the odds fall to something like 1 in 6,000 to 15,000. From “just don’t lower my living standard” to “a young, single, self-made multimillionaire,” the probability drops roughly six-thousandfold. That is no longer mate selection; it is close to buying a lottery ticket.
It helps to see where such thresholds sit in the real distribution. China’s median disposable income is about ¥2,900 a month, and even the top fifth of the population averages only about ¥8,200. A standard set in the far tail of that distribution is not so much a high bar as a decision to leave the market – because no feasible amount of searching can supply it.
Why it looks like a dead-end
For a career-focused woman, the filters stack fastest. She tends to want a partner older and higher-earning; He in turn wants children. And because she had spent her earlier years focusing on her career, she was no longer in her prime when she started looking for a partner. Thus the requirements compound – older, higher-earning, and willing to accept a partner who is herself no longer young. Add the biological fact that the prime career-building years overlap heavily with the prime fertility years, and four constraints pull tight at once. For a woman who wants both a serious career and a family, it can look like an unsolvable problem.
But that conclusion rests on an unstated premise: that the norm behind the whole calculation – the man must earn more, and the woman carries the caregiving – is a fixed constant. It isn’t. And the cleanest way to see that is to watch the same situation play out under different norms.
What the data actually shows
In the United States, “college-educated women are the hardest to marry” was once conventional wisdom. Over recent decades that gap has reversed: highly educated American women now marry at least as readily as less-educated women, and divorce markedly less (Isen and Stevenson, 2010). Northern and Western Europe are narrowing the same way (Bertrand and colleagues, across 23 countries). Yet in East Asia – China, Japan, Korea – and Southern Europe, the gap persists or widens, closely following the “leftover women” pattern.
Here is the logic that matters. The same input – high female achievement, plus identical human biology – produces “marries more, divorces less” in America and “leftover women” in East Asia. If success itself, or biology, were the cause, the American reversal could not have happened. Since it did, the cause can only be the one thing that systematically differs between the two regions: gender norms. The sociologist Yue Qian, who studies China’s case, finds that highly educated Chinese women gain little from a “male breadwinner” marriage, and so rationally delay or forgo it. It is a personal-level symptom of what the demographer Peter McDonald calls the gender-equity mismatch: the public sphere has equalised, the private sphere has not caught up. What traps the successful woman is not her success. It is a norm that has not been updated.
The levers that move the odds
Once the bottleneck is norms and standards rather than biology, the levers become clear – and each is evidence-backed:
- Relax “he must earn more.” This is the highest-return, lowest-cost move. It is simultaneously the filter that shrinks the pool most (1 in 16 to 1 in 87) and the one that least predicts whether a marriage lasts or satisfies – the reliable predictors in the literature are dependability, kindness and shared values, not who out-earns whom. Dropping it is not “settling”; it is removing an expensive constraint that buys nothing.
- Use money to outsource the double burden. What makes the problem look unsolvable is the assumption that she both builds a career and does all the childcare. High income can simply buy that away – hired childcare, outsourced housework – dissolving the “career or children” choice.
- Screen for private-sphere equality, not a bigger salary. In a norm-lagging environment, the trait worth filtering for is a partner who genuinely wants an equal, earning partner and will share the caregiving – who is, for a career-focused woman, the better and more durable match anyway.
- Aim at one person, not a ratio. “1 in 87” is frightening as a population statistic but irrelevant to a one-to-one search; high achievers cluster in a few cities and industries, and a small fraction of a dense enough pool still yields matches.
The one genuinely hard residual, which money cannot fully solve, is the fertility window: fertility declines with age (more steeply after 35), and that decline overlaps with the career-building years. The available lever is to decouple the timing of children from the timing of marriage – egg freezing and assisted reproduction do this partially (more effective the earlier it’s done, and never guaranteed) – or to start looking for a partner earlier, during the climb rather than after it. This is a real trade-off that can be eased at a cost, not erased.
Not “no solution” – just “not all at once”
So is it true that successful women find it harder to marry? The honest answer is that the problem is not unsolvable – it is only impossible to have everything simultaneously. What is genuinely unsolvable is wanting, all at once: a top-tier career, children at the biologically ideal moment, a partner who out-earns you, never outsourcing care, and a society whose norms lag behind. You can let go of one. The data points clearly to which one costs least: not the career, not the children, but the requirement that a partner earn more – because it is both the sharpest shrinker of the pool and the least connected to whether a marriage is any good.
And there is a final, uncomfortable point that dissolves a common confusion. In outcome, “I want to marry but my standards make it near-impossible” and “I don’t want to marry” end the same way – both leave you unmarried. But in cost they are opposites: the person who genuinely doesn’t want it pays nothing, while the person who wants it yet draws the line out in the statistical tail keeps rejecting people who would have suited her, and carries years of frustration, for the identical empty result. The question was never “should I settle.” It was whether the price of each standard has ever been counted – because the pen drawing that line has been in her own hand the whole time.
Previously in this series: Why have both marriage and birth rates declined?
Next in this series: How did fashion become essentially worthless?
