When a Wife Earns More

A new working paper (abstract; PDF) by Marianne Bertrand, Jessica Pan, and Emir Kamenica looks at gender identity and its effect on household income. Their findings will depress anyone concerned with gender equality. Here’s the abstract:

We examine causes and consequences of relative income within households. We establish that gender identity – in particular, an aversion to the wife earning more than the husband – impacts marriage formation, the wife’s labor force participation, the wife’s income conditional on working, marriage satisfaction, likelihood of divorce, and the division of home production. The distribution of the share of household income earned by the wife exhibits a sharp cliff at 0.5, which suggests that a couple is less willing to match if her income exceeds his. Within marriage markets, when a randomly chosen woman becomes more likely to earn more than a randomly chosen man, marriage rates decline. Within couples, if the wife’s potential income (based on her demographics) is likely to exceed the husband’s, the wife is less likely to be in the labor force and earns less than her potential if she does work. Couples where the wife earns more than the husband are less satisfied with their marriage and are more likely to divorce. Finally, based on time use surveys, the gender gap in non-market work is larger if the wife earns more than the husband.


Correlation does not equal causation. Opening umbrellas does not cause rain. "Couples where the wife earns more than the husband are less satisfied with their marriage and are more likely to divorce." Unless the study started paying random women more money (something that no survey has the budget for, if done in the united states), this proves absolutely nothing. It could be that when couples that fight more, a woman focuses more at work and then makes more money. Junk science.

Not sure you understand

I'm not sure you're understanding the paper, or indeed, the way academic economic language works in general. You're right that correlation is NOT causality, and that is not at all claimed by the paper! The statement that "couples where the wife earns more than the husband are less satisfied with their marriage and are more likely to divorce" says NOTHING about causal relationships, only statistical ones.

Of course, the explanation you are proposing is easily testable: if fighting couples generally tend to have higher female income, then there may be water in what you arguing. But anecdotally, there seem to be many couples with lots of fighting where the man does earn more; I would be surprised if overall statistics found otherwise.


The wording by freakenomics falsely implies causality. Your brain makes an implication from a factually apparently neutral statement. Reword it to see that --"when in less satisfying relationships women earn more". There's an automatic human false implication that the writer is trying to push.


Can someone explain why this would be true? "Finally, based on time use surveys, the gender gap in non-market work is larger if the wife earns more than the husband."

It's seems like if women earn more they would have less time for non-market work.


It's in the paper - 'This result runs counter to standard models of the division of labor within the household (e.g., Becker 1973), which predict a negative relationship between the wife's share of market income and her relative contribution to home production activities. One explanation for the observed pattern is that, in couples where the wife earns more than the husband, the "threatening" wife takes on a greater share of housework so as to assuage the "threatened" husband's unease with the situation. The wife, of course, may ultimately get tired of working this "second shift" (Hochschild and Machung 1989), which could be one of the mechanisms behind our results on divorce.' Bottom of page 4.


This may be disappointing to some but I hardly see how it should be surprising to anyone. There have been studies / surveys which recently support this attitude regarding gender roles, even in the feminist media circuit they are universally panned. For example, the unwillingness of women to date unemployed men (while the reverse was not true).

Ultimately under either progressive/feminist frames of thought, or traditionalist frames of thought, the idea of a man making less money than a woman is something that reflects poorly on him. Under traditionalist approach it's obvious why: Because the man is "supposed" to support the family. Under a progressive/feminist approach it's a little more nuanced. But, under feminism the idea of Male Privilege is rife, and the myth that men earn significantly more than women for the same work is pervasive. Combining these two things you have an environment where male under-achievement will be greatly looked down upon. After all they get paid more for the same work AND have all the privilege to get them up into the upper echelons of employment, according to feminist theory.

So, under either popular framework there will be resentment / inadequacy for a man who makes less than a woman.

This is only going to get worse as time goes on, as well. More money is spent on education for women, and support systems for men and the networking "advantage" they had in terms of employment is rapidly disappearing with male education going down the toilet (female degrees : male degrees is around 4:3 now). Expect lower rates of coupling/marriage, higher rates of divorce, and higher suicide rates for males going forward. Also expect new feminist media in the tradition of Hana Rosin and The End of Men to further put the "failure" of men on their shoulders.



Earning differential does not define marriage or divorce; values differences do. My husband and I celebrated 44 years of marriage this year. I currently earn about 5 times his salary. He's moved and switched jobs twice for my career. It's ok. We share the same social, family, and in large part personal values. I know several couples where this holds true. I agree .... junk science.


One anecdote doesn't invalidate a study or make it "junk science". You are the exception, but most people are the rule, assuming the research is thorough and peer-reviewed.


I'd be curious to find out, based on studies, in households where the woman is the breadwinner and the man stays home: (1) who does what housework, how much does each person do, and how does it compare to households where the the man works and the woman stays home? and (2) if kids are involved, how does the burden of parenting break down? IE, who provides the most care to the children? The working mom? The stay-at-home mom? The working dad? The stay-at-home dad?

Additionally, it would be fascinating to compare opposite sex couple households with same sex couple households.

Rob Brooks

Wonderful paper. Very interesting indeed.

Folks who are interested in this paper might also be interested in the "Cooperation and Conflict in the Family" conference - to be held at UNSW in Sydney next February. I hope you don't mind me posting the link here



blah. generalised everything. boring.


Unfortunately the authors reveal their bias from the very beginning by posing only two alternatives; that men and women both have equal preference for a partner with higher income, or that men prefer a partner with lower income. However, the evidence that they claim supports the latter view can be equally well explained by a third alternative; that women have a strong preference for a partner with higher income, while men have a weak preference or are neutral regarding partner income.

There is in fact strong evidence from previous studies that this is the case, yet the authors refuse to even consider it by discounting data regarding each partner's marital satisfaction (one wonders if they did this only after analyzing the data and finding the conclusion to be inconvenient for their hypothesis).

Finally, regarding the findings on non-market work, this is presented with no consideration of mitigating factors. There is a presumption that the partner with the higher income must also be working more hours. Do male partners with lower incomes work longer hours to increase their income, commute further for a higher paying job, have jobs with less flexible schedules?

I don't totally discount the paper's thesis. There may in fact be a subgroup of men who prefer partners with lower incomes. However this paper does not provide evidence for this, and I suspect that if such a group exists it consists primarily of older men with very high incomes. While the "average" man might not make income the primary attribute in choosing a partner I cannot imagine that he would view higher income as anything other that a positive in choosing between two otherwise similar women.