A study of 544,000 couples across 29 countries confirmed that relationships fail more often when women earn more, then ruled out almost every explanation researchers assumed was responsible

For decades, the explanation has felt obvious. When a woman earns more than her male partner, the relationship faces a specific kind of pressure: the gap between what society expects and what the couple actually is. Men are supposed to be providers. When they are not, the reasoning goes, something strains. Both partners feel the weight of violated expectations. The man’s sense of identity is threatened. The woman compensates by taking on more at home. The relationship destabilizes.

This explanation has shaped not just research but cultural conversation for years. It is the premise behind countless articles about breadwinner wives. It is the assumption embedded in the phrase “traditional gender norms” every time a researcher tries to explain why female-higher-earning couples divorce at elevated rates. It feels so self-evidently correct that most people who encounter the statistic nod and move on.

A new study of 544,911 couples across 29 countries has confirmed the statistic. Female-breadwinner couples are 36 percent more likely to separate than male-breadwinner couples. The finding held across nearly every high-income nation on earth. But when the researchers set out to identify which explanation best accounted for this pattern, almost none of the ones everyone assumed were responsible survived contact with the data.

The scope of the study

Led by Pilar Gonalons-Pons of the University of Pennsylvania and Allison Dunatchik, the study drew on harmonized panel data from large international household surveys covering 29 high-income countries between 2004 and 2020. The dataset included 437,102 married couples and 107,809 cohabiting couples, making it one of the largest cross-national studies of relationship dissolution ever conducted.

The scale was deliberate. Testing whether traditional gender norms drive the female-breadwinner separation effect requires variation across countries with meaningfully different cultural attitudes toward gender. A study conducted in one country, or even a handful, cannot distinguish between a pattern specific to particular cultures and one that holds universally. With 29 countries spanning different legal systems, welfare states, cultural traditions, and labor markets, the researchers had what they needed: enough variation to see whether the effect changed in the ways it should if the leading explanations were correct.

It did not.

What the leading theories predicted and what the data showed

The gender norms explanation predicts a specific pattern: the female-breadwinner separation effect should be stronger in countries where traditional attitudes toward gender roles are more prevalent. If the mechanism is that couples feel the strain of violating cultural expectations, then the strain should be higher where those expectations are strongest.

The researchers tested this directly. They measured cultural attitudes toward gender roles across their 29 countries and looked for the interaction. Countries with more traditional gender attitudes did not show a stronger association between female breadwinning and separation. Countries with more egalitarian attitudes did not show a weaker one. The effect was essentially the same across the cultural spectrum.

The financial independence explanation — that women who earn more are simply better equipped to leave relationships they want to leave, and that the higher separation rate reflects greater exit capacity rather than greater relationship strain — also failed to account for the data. If financial independence were driving the pattern, it should be stronger in countries where women face greater barriers to economic independence and weaker where women already have strong labor market positions. No such variation emerged.

The economic similarity hypothesis, which proposes that people seek partners with similar socioeconomic backgrounds and that female-breadwinner couples are inherently mismatched, also received little support. Nor did explanations focused on men’s relative economic disadvantage within the relationship.

Four of the most prominent theories in the literature, all of which have generated decades of research and policy discussion, arrived at the same destination: not ruled out entirely, but unable to account for the pattern they were supposed to explain.

What the data did support

One explanation did find meaningful support, and it pointed in a direction that reframes the entire conversation.

The researchers found that the female-breadwinner separation effect was substantially stronger among couples with children than among childless couples. Female-breadwinner couples with children were 49 percent more likely to separate than their male-breadwinner counterparts. Female-breadwinner couples without children were 23 percent more likely to separate. The difference between those two numbers, 26 percentage points, is not noise. It is a signal about where the actual mechanism lives.

The pattern points toward what the researchers call work-family conflict: the structural collision between the demands of paid employment and the demands of family care. When women earn more than their partners, they are, by definition, not scaling back their careers after having children at the same rate that women in more traditional earning arrangements do. The domestic labor that children require, which research consistently shows continues to fall disproportionately on women regardless of employment status, then accumulates on top of a full professional workload.

“Work-family conflict might be particularly intense when women do not scale back from paid work after having children and male partners do not sufficiently contribute to domestic labor,” the authors write.

The mechanism is not, on this account, about how men feel about being out-earned. It is about the concrete, daily reality of who does what when a household has both children and two demanding careers, and one of those careers belongs to the partner who society and most domestic labor research still expects to shoulder the invisible work of family life.

Why the distinction matters

If traditional gender norms were the primary driver of the female-breadwinner separation effect, the policy implication would be to change attitudes. More egalitarian cultural beliefs would gradually erode the problem as younger generations replaced older ones. There is evidence that this is already happening in many countries, and the conventional wisdom has assumed the separation effect would follow.

But if work-family conflict is the primary driver, the problem is structural rather than attitudinal. It persists even in countries where people hold egalitarian beliefs about gender, because holding egalitarian beliefs does not automatically produce egalitarian behavior inside households with small children. The couple who are entirely comfortable with the woman earning more may still find that the distribution of domestic labor does not reflect that comfort, not because they are hypocrites, but because they are embedded in labor markets, care systems, and workplace cultures that were built for a model of family life that no longer describes the majority of households in high-income countries.

Parental leave policies that incentivize fathers to take time off. Workplace cultures that do not penalize reduced hours or flexibility for caregiving. Affordable, high-quality childcare that reduces the domestic coordination burden on whichever parent manages it. These are not primarily cultural interventions. They are structural ones, and the data suggests they are what the female-breadwinner separation effect is actually asking for.

What the study cannot tell us

The researchers are explicit about the limits of what 544,000 couples across 29 countries can establish. The dataset lacks information on relationship duration, which is relevant because separation risk changes over time. Large international surveys cannot capture relationship quality, communication, compatibility, or the countless individual factors that determine whether a specific couple stays together. The study is limited to different-sex couples in high-income countries, meaning its findings may not generalize to same-sex couples or lower-income settings.

The 36 percent figure is a population-level association, not a prediction about any individual relationship. Many female-breadwinner couples are highly stable. Many male-breadwinner couples separate. What the study identifies is a pattern across nearly half a million couples that is real, consistent, and almost certainly not caused by what most people have spent decades assuming it was caused by.

Source

Pilar Gonalons-Pons, Allison Dunatchik. “Women’s Socioeconomic Advantage Over Their Partners and Relationship Dissolution: A 29-Country Study.” Journal of Marriage and Family, 2026.
DOI: 10.1111/jomf.70081