What happens to a preference for a partner who out-earns you once you are the one earning the most? For a growing number of women with graduate degrees and senior titles, that question stopped being hypothetical. The preference was inherited from a period when it described most of the available options. It now describes a shrinking fraction of them, and the women holding onto it are the ones with the least room left at the top of the ladder.
The Preference and Its Pedigree
The pattern is well established in the research literature. David Buss surveyed 10,047 people across 37 cultural samples in 33 countries, and women in 36 of those 37 samples rated good financial prospects higher than men did. Men in the same samples preferred younger partners while women preferred older ones. The result held in Sweden and Norway as firmly as in Iran, which is why it survived decades of arguments that the preference was purely a product of local custom.
That consistency makes the preference a durable statistical tendency, and a tendency is something a person can examine rather than obey. The women most affected are the ones who were raised on the tendency, worked hard enough to invalidate its premise, and never revisited the criterion.
Marriage Outcomes Against Preference
Stated preference and actual pairing diverge in interesting ways here. Between 2019 and 2023, most women in the four lowest income quintiles did marry higher-earning men, which is the expected result. In the highest quintile the figure was 43%, a minority, though a much larger minority than a purely income-blind sorting would produce.
The other half of that finding gets less attention. If 43% of top-quintile women married up, the remaining 57% did not, and those marriages are the ordinary outcome for anyone whose income puts a hard ceiling on how far up the ladder actually goes.
Venues, Criteria and the Search Itself
Criteria come before venues, though most people work in the opposite order. A woman who has not settled what she is weighting will weight income by default, because income is the number easiest to read at a distance.
Women who set out to meet rich guys usually solve the venue problem quickly. The harder part is settling in advance what will actually matter once someone is standing there.
The Arithmetic of a Shrinking Pool
The supply problem is measurable. Pew Research found 47% of American women aged 25 to 34 hold a bachelor’s degree against 37% of men in the same bracket, and women have been outpacing men in college completion in every major racial and ethnic group. Women have outnumbered men among college graduates since 2015.
The pipeline behind those figures is wider still. Women make up close to 60% of undergraduate enrollment, they earned 58.5% of bachelor’s degrees in 2021 and 2022, and their advantage in college enrollment and graduation has widened rather than closed. A woman with a professional degree who requires a partner with equal or greater credentials is drawing from a group that is smaller than her own by a margin of roughly 10 percentage points before any other criterion is applied. Add an age band, a location and a personality requirement, and the remaining set becomes very small.
What the Filter Screens Out
The filter is expensive in a way that is easy to miss, because the men it removes are not obviously deficient. NPR reported that even as women outnumber men in colleges, their later earnings have stayed comparatively flat, which means credentials and income have come apart from each other in both directions. A man in a skilled trade may out-earn a man with a master’s degree, and a man with a master’s degree may earn less than the woman assessing him. Neither fact is legible from a profile or a first conversation.
What the criterion screens out most reliably is men who are steady, present and willing to organize their life around a partner whose career moves first. Those men exist in reasonable numbers. They tend to be filtered at the top of the funnel, before anyone has learned anything about them, on a proxy that was never a very good measure of what makes a household work.
Status Signals Beyond Income
Income is one status signal among several, and it is the one that has become least reliable. Women in the lower income brackets still tend to marry higher earners, though the spousal gap keeps narrowing at every level. The share of opposite-sex American marriages with a breadwinner wife has moved from about 5% in 1972 to roughly 16% today, and a combined 45% of wives now earn the same as or more than their husbands, close to triple the 1972 share.
Competence, social standing, physical presence and the ability to hold a difficult conversation are all status signals, and none of them appears on a pay stub. A partner who is respected in his field, calm under pressure and good with her family is delivering most of what the income proxy was standing in for, and he is doing it in a form that a first meeting can actually assess. A woman who has already secured her own material position is in an unusual spot, because she can afford to weight the signals that predict a workable partnership over the one that predicted material security for her grandmother. Very few people take that option, mostly because the criterion feels like taste, and taste is rarely audited.
A Preference Worth Re-Examining
The uncomfortable part of this is that the preference is not irrational. It tracks a historical condition in which a partner’s resources decided how well a woman’s children fared, and the Buss samples show how widely that tuning still holds. It does not switch off because a woman made partner at 34. What has changed is that the preference now costs the women who hold it most strongly the most, and it costs them in the currency of years. A criterion that reliably eliminates 80% of the men she meets has stopped functioning as a standard and started functioning as a search set to return nothing. Only the person who set it can change the parameters.