How housework affects birth rates

When women think their share of household labor is unfair, they have fewer kids

I love a good argument, especially when someone smart takes the time to engage thoughtfully with claims that I’ve made. Therefore, I was delighted when Lyman Stone, a demographer I respect, published an article on the Institute of Family Studies website that engaged with my Slate article and accompanying Substack explainer on the link between birth rates and male housework participation. Stone’s article was titled “Men Doing More Housework Won’t Raise Fertility,” but the data he presents tell a more nuanced story than the headline. In fact, I actually think his article does a great job of supporting and extending some of the claims I made in my original piece.

In the spirit of furthering a friendly argument, here’s my take on Stone’s piece.

First, to briefly summarize my original piece: I argued that men’s contributions to housework and childcare were associated with higher birth rates in wealthy democracies (OECD countries). Much pronatalist discourse blames women’s advancement for falling birthrates. Take the viral twitter post suggesting that female education is the problem. Rolling back the gains of feminism is an implicit goal of the MAGA pronatalism push.

However, men’s choices also matter for birth rates. For all the handwringing about working moms, there has been little discussion of male contributions to household labor, even though it is obvious to me, and every mom I know, that women’s reluctance to have children is often tied to her reluctance to carry most of the extra unpaid labor. In my piece, I compared TFRs (total fertility rates) with men’s household time use (male participation in unpaid domestic work relative to women) within wealthy democracies (OECD countries), and found moderate-to-high positive correlations. Male domestic labor helped explain the difference between lower-TFR countries like South Korea and Italy, and higher-TFR countries like Denmark, Norway, Sweden, or Finland.

Stone argues that the cross-sectional approach (that is, looking at unpaid labor + TFRs sampled at the same time across multiple countries) fails to demonstrate causality. As he writes:

There are many reasons fertility or child care arrangements might vary. Do Swedish women have more babies because Swedish men do more housework? Or do Swedish men do more housework because Swedish women have more babies? Or is there some other fact about Swedish culture causing more babies and more men to do housework?

He is correct, and I agree that these are multidetermined phenomena. I think we can safely throw out the second possibility (“do Swedish men do more housework because Swedish women have more babies?”) when we focus on the share of unpaid care labor that men do relative to their partner, not their overall hours performed on care labor. The total domestic workload rises when people have kids, but so does gender inequity. Even when couples are egalitarian before birth, women do relatively more after having a child. There’s no reason, therefore, that we’d expect men’s share of domestic labor to increase with more babies (unless, for some reason, we’re talking only about domestic labor that excludes infant care). But the final possibility Stone mentions (“Is there some other fact about Swedish culture?”) strikes me as probable. In fact, I concluded my original article with one such third-variable explanation, the societal value of care labor. Cultures in which men participate more in care also tend to be cultures with greater female political leadership, stronger safety nets, and higher investments in early childhood. As I wrote in my piece, “In these countries where male participation in parenting is high, there is also strong public support for caregiving and an emphasis on healthy child development. Childcare is not seen as a personal favor that women do for their families, but as a responsibility shared by all adults.” Male contributions to housework are a reflection, not solely a driver, of pro-child cultural values.

In any case, Stone tackles the causality issue by doing longitudinal analyses of trends over time within countries, rather than across countries. Whether you prefer a cross-sectional or longitudinal approach depends on whether you are more interested in comparing countries’ current situations at a particular snapshot in time, or whether you want to draw causal inferences through repeated measurement of the same phenomena. Both approaches are valid, but the longitudinal design allows for stronger claims.

Stone uses four sources of data to test longitudinal models. First, he looked at data form the International Social Survey Program (ISSP), which measured four chores multiple times from 1994 to 2022: laundry, cooking, caring for sick family, and grocery shopping. Second, ISSP also measured self-reported hours of domestic work at three timepoints between 2002 and 2022. Third, a few countries have actual time-diary data, which is the gold standard. For each of these, Stone looked at the share of domestic work performed by partnered men between the ages of 18 and 54. Finally, Stone looked at respondent perceptions of fairness from the ISSP.

So what did he find? First, across the first three measures, he partially replicated the claim that male housework participation is associated with birth rates cross-sectionally. For example, when looking at married men’s share of reported time on household work, he concluded, “Across countries, it looks like where men do more housework, fertility is higher.” When it comes to time diary data, he writes, “If you squint…you could perhaps convince yourself that in countries where men do more housework, fertility is higher: Canada and Finland are much higher than Korea or Italy.”

However, what happens when we look at trends over time? They are quite a bit messier. For some countries, the lines have an upward slope, suggesting that when men increase their domestic work over time within a particular country, that country’s birth rates rise as well. For the four-chores data, that’s true for Spain and the Netherlands; for self-reported housework data, there are positively-sloped lines for Hungary, France, and Germany; and for time diary data, there is only a positively sloped line for Finland. But there are just as many countries in which the line is flat or points down, suggesting that increases in male housework have a neutral or negative effect on birth rates within countries.

Evidence for a mixed relationship does not necessarily prove the null hypothesis that no relationship exists. In fact, these mixed results suggest few different possibilities. Maybe there’s a non-linear association, or maybe there is a specific “dose” of male housework contribution that makes a difference. Stone alludes to this in discussing the four-chores data: “The highest fertility countries actually have “moderate” levels of men’s housework: men’s housework shares above 35% are negatively related to fertility. Second, it is true that many very low fertility countries have very low rates of men helping out at home.”

In other words, it might matter when men are doing more than a certain minimum, but there are diminishing returns past a certain threshold. This seems intuitively correct to me. When men are helping out a lot, that might reflect periods of national economic uncertainty or male unemployment that could be dampening fertility. That was the case in the wake of the 2008-2009 recession. But when men do less than, say, 25%, women don’t want to form families with them. For example, consider Japan. Stone points out that “Japanese men increased their housework share from 7% to 23% between 2002 and 2022. Fertility stayed about stable.” Although 7% to 23% is a significant increase, it occurred alongside an rise in Japanese women’s employment. As of 2024, 80% of Japanese mothers are employed outside the home, a record high number. If those mothers are also doing 77% of the household labor, that’s a bad deal for them, even if it reflects a considerable uptick in men’s domestic contribution. Maybe there is a “sweet spot” in men’s household work that is optimal for fertility, and upward and downward shifts towards that sweet spot will boost birth rates. Interestingly, Eve Rodsky found something similar when she tested the Fair Play deck. Couples did not need to reach a 50-50 split in order to be happy with their domestic arrangement. As long as men were holding at least 21 cards – which, depending on how many cards the couple has in play, might translate into something like 25-30% of the total workload – couples tended to think their division was fair.

One drawback to Stone’s analyses, which he readily admits, is that the most complete data he has (the four chores) is also the least comprehensive. Tasks like grocery shopping and laundry may be a decent proxy for general involvement in the household, but these four chores do not include routine childcare. “Caring for sick family” is the closest approximation for care labor, but it is also a low-base-rate measure that depends on how often people in the household are getting sick, which could be confounded with lots of other variables. Using these four chores is sort of like calculating the calories in your grilled cheese and tomato soup by looking at the back of the tomato soup can. Tomato soup is part of your lunch, but it’s not the main meal!

That brings me to the fourth measure that Stone uses in his analysis, one that he dismisses as less meaningful but that, in my view, might be the most useful metric of all. That is another question from the ISSP: whether respondents felt that they were doing more or less than their fair share of housework. Importantly, this measure is only very weakly correlated with the housework time measures described above. But, when you look at fertility, Stone writes, “…Scholars focused on housework just might have a good point: while objective housework distributions don’t matter very much for fertility or for partners’ sense of fairness, subjective views of housework distributions may still matter!” Within countries, the time trends show a negative slope; that is, when women’s perceptions of unfairness increase, fertility falls. This is, as Stone writes, “a fairly large observed effect and is around the threshold for statistical significance.”

This is an exciting finding! As Stone writes, “The central intuition that when women feel their burden at home is unfair, they have fewer children is true.” But Stone and I derive totally different conclusions from it.

Because women’s perceptions of unfairness are not well correlated with men’s reported housework share, Stone asserts that “Unfair home burdens don’t really have much to do with the amount of work men do at home.” He writes “The takeaway from all of this is simple: pressuring men to help out more at home won’t boost fertility.” I disagree with this takeaway. In fact, I believe that subjective perceptions of unfairness might actually be our best measure of men’s relative contributions, and that there is clear evidence that achieving a mutually satisfying division of labor will support family formation.

Stone’s finding that women’s sense of fairness is not strongly correlated with men’s self-reported share of housework is no big shocker. It is tricky to capture domestic workload using retrospective self-report. How much time did you spend on housework last year? When asked this question, many people over- or under-estimate their totals. In fact, there’s a whole research literature on couples’ perceptions of their relative housework contributions, suggesting that both spouses tend to inflate their share. It’s the domestic-workload version of Lake Wobegon, where all the children are above average.

Partners base their division of workload expectations on multiple variables. If I’m working part-time and earning a third of the household income, I might be totally happy doing 70 or 75% of the housework at home. If I’m the primary breadwinner working 80 hours a week, that 70% feels deeply unfair. My “fairness” rating is not solely derived from my husbands’ housework share. That doesn’t mean that fairness isn’t meaningfully linked to his contributions at home! It just means that his contributions need to be weighed against other household dynamics.

Finally, there’s another reason why subjective unfairness may approximate the domestic workload better than self-reported housework time, and, conveniently for me, I already study it. My lab has examined the division of cognitive household labor, also called the “mental load.” Chores like grocery shopping or laundry have an execution dimension – the person who drives to the store, gets everything on the list, and then hauls the bags home, or the person who throws clothes and detergent into the washing machine – and a planning dimension – keeping tabs on everyone’s food preferences, tracking when supplies run out, monitoring everyone’s laundry hamper, making sure the detergent is in stock, and knowing that your son’s soccer uniform is filthy but he has a game tomorrow morning. Mental load has also been called “invisible labor” because it frequently goes unacknowledged or even unnoticed by other members of the household. Since it “runs in the background” while people engage in different tasks, it’s hard to measure.

Our lab did a study, led by my graduate student Lizzie Aviv and published in the Archives of Women’s Mental Health, to assess the cognitive dimension of household labor. Eve Rodsky collaborated on this project with us, and we adapted 30 cards from her Fair Play deck into a survey that we gave to 322 mothers of preschool-aged children. Each card represented a common household task. For each card, we asked participants to rate their share of both task execution and task planning. For 28 out of the 30 tasks, moms said they did a larger share of execution. But for every card in the deck, moms said that their share of the cognitive work exceeded their share of the execution. In other words, “mental load” was more unevenly distributed than task execution. Moreover, women’s division of cognitive household labor was linked with their reports of stress, burnout, depression, overall mental health problems, and relationship dissatisfaction. The mental load was both more gendered, and more psychologically taxing, than the more measurable aspects of housework.

It’s possible to have a partner who is objectively helpful around the house but still treats their spouse like the default parent or cruise director of the household. Some partners enjoy having this “chief executive” role at home, but our data suggest that many women are unhappy carrying all the mental load. Women’s perceived fairness ratings may reflect the mental load as well as the visible load that women shoulder within relationships. (For an even more visceral perspective on the mental load, I recently stumbled across this excellent essay by Diana Fox Tilson, LICSW. If you’re tempted to say that mental load is just an artifact of women overthinking household work and caring too much about silly details, read about her experience being raised by a single father).

So what does this mean for the birth rate? Here’s Stone’s take:

For fertility to rise, what countries need is not for one sex or the other to juggle their duties differently, but for society at large to take some of the balls out of the air so they simply don’t have to juggle as much, and also for cultural narratives to push back against framing of family life as an arena of power dynamics, competition, and gendered struggle. How much help women get at home doesn’t impact society-wide judgments of fairness. But how much society around them preaches that men are sexist might.

I completely agree that society should take some of the balls out of the air and reduce the stress on parents. Yes! Let’s discard intensive parenting expectations and champion a more supportive societal infrastructure for raising kids. However, I disagree we must also “push back against framing of family life as an arena of power dynamics, competition, and gendered struggle,” or that women’s perception of unfairness is due to “how much society around them preaches that men are sexist.” If I’m working the same hours as my spouse, doing most of the housework, and carrying all the mental load, and I think that balance is unfair, is that because society has taught me that men are sexist, or because it actually is unfair? In fact, within the rising numbers of couples who earn equal incomes, women spend about two hours more on housework and 2.5 hours more on childcare than their partners, and men report about three more hours a week of leisure. Of course family life is an arena of power dynamics. It has always been an arena of power dynamics. The majority of mothers are in the paid workforce. Why should women settle for a balance of housework that they consider unfair? The fact is that many women are not settling for this balance, and in fact they are registering their dissatisfaction through expressing a reluctance to form families, which is the whole reason that both Stone and I are writing about this topic in the first place.

Stone does not like the idea of “pressuring men to help out more at home.” Personally, I’m also not a fan of “pressuring men,” but I do favor policies that give men more dedicated time at home with kids (paternity leave, part-time-friendly work schedules that allow for things like the Danish papadag) and for cultural change to elevate the perceived societal value of childcare. These are good things that benefit men, not pressure tactics!

The Institute for Family Studies is a right-leaning organization that advocates for traditional family norms, so I’m not surprised that Stone’s piece has the framing or headline that it does. (Stone and I also had a lengthy back-and-forth via DM about some of these very same issues). Two researchers can look at the same analyses and draw different conclusions. My takeaway is two-fold: men’s share of household labor is associated cross-sectionally with birth rates across countries, and women’s subjective sense of unfairness predicts lower birth rates over time. Both findings run against the prevailing pronatalist wisdom that simply reinstating traditional gender roles will boost the birth rate. In fact, it suggests the exact opposite of this frequently echoed assumption. If we want to raise the birth rate, we should care about whether women feel like they are getting a fair deal within their marriages.