Hal Weitzman: Women in the United States earn an average of about 82% of what men earn. That gap has barely narrowed in the past 20 years.
Among the many causes, to what extent can we explain the gender pay gap by what happens not in the workplace but at home? Do parents act in ways that challenge or reinforce the economics of gender? How do children view gender roles? And what do we need to do to close the gender pay gap?
Welcome to the Chicago Booth Review podcast, where we bring you groundbreaking academic research in a clear and straightforward way. I’m Hal Weitzman, and in this episode we’re looking at research about the economics of gender, from the behavior of parents to the attitudes of children, spouses, and bosses. As we’ll hear, achieving gender equality is not just about women.
In many cases, women have been held back by not being given the same educational opportunities. That often has to do with parents’ decisions about spending. So how does the gender of the child affect those decisions? Chicago Booth’s Rebecca Dizon-Ross and Princeton’s Seema Jayachandran conducted research on that question that we covered in the Winter 2022/2023 issue of Chicago Booth Review’s print magazine under the headline “Some Fathers Are Less Willing to Spend on Daughters than on Sons.” The article was written by our contributor Kasandra Brabaw, and it’s read by Julie Granada-Honeycutt.
Reader: Gender plays a big role in how much parents are willing to spend on their children, according to Chicago Booth’s Rebecca Dizon-Ross and Princeton’s Seema Jayachandran. Daughters tend to lose out when their fathers are in charge of spending decisions, they find, and the researchers speculate that it’s mostly because the dads just like boys better.
The results, based on a series of 2013 surveys in Uganda, demonstrate that mothers there tend to be more evenhanded. This could have important implications for policy making, suggesting that advances in gender equality will be self-reinforcing, the researchers argue. “As women gain more say in household decision-making, household spending on daughters may increase, producing more gender equality in the next generation,” they write. “This virtuous cycle could help to close the gender gaps in schooling and health care that are pervasive in developing countries.”
The researchers recruited more than 700 two-parent households, each with a mother and father, in rural Uganda to participate in surveys measuring the parents’ inclination to spend on both crucial (education- and health-related) and “fun” items for their children who were enrolled in grades four through six. Some households also had younger children.
In the first round of data collection, one parent from each family was asked a series of questions about household composition, family background, and each parent’s financial situation. Respondents also gave their thoughts on education in relation to gender, such as whether they believe that it’s useless to send girls to secondary school because they will eventually marry, or that boys will become better farmers if they attend school.
They were asked directly which parent was more caring toward their children and were then given a “willingness to spend” survey covering a variety of items, of about a dozen different prices, for their daughters and sons. A second survey round asked similar questions—with some variation—of the other parent in each household.
The first survey assessed the parents’ willingness to spend on three “human capital” items—a practice exam administered by the child’s school, a deworming medication, and rubber-soled shoes. The second survey added a school workbook in the human-capital category, along with two fun items: candy and a ball.
Analyzing responses from both surveys, the researchers find that mothers would generally spend close to the same amount on sons and daughters, with a willingness to spend marginally more on health and education for their daughters. Fathers overall had a 6 percent lower willingness to spend on daughters than on sons.
The researchers probed for the reasons behind this, exploring whether the parents’ willingness to spend reflected a return-on-investment approach (which child, son or daughter, would provide old-age care) or simple gender preference. They find that the preference-based explanation is likely—that fathers care more about sons than daughters, and mothers don’t display that tendency.
The researchers argue that if the disparities appeared only in human-capital spending, they might reflect differing perceptions about return on investment in a child. But fathers were not only less willing to spend on daughters in the human-capital category but also for items that might bring them joy. Mothers were willing to spend slightly more on human-capital items for their daughters but would spend equally on them for the candy and the ball.
Moreover, in about half of households, both parents said that the mother was the more loving parent. And in those households, the pattern of fathers spending less on daughters was particularly strong. Dizon-Ross says this provides more evidence that gender preference is at work by making the argument “that our mother-father differences in girl-boy spending are driven by the households where both parents think the mothers love more, which might proxy for loving girls more.”
Even if mothers are willing to spend more on health and education for their daughters out of gender preference, the researchers observe, men in poorer countries such as Uganda generally control more financial resources than women do, so daughters are more likely to lose out.
Hal Weitzman: If educational patterns are formed when we are children, so are many of our views. Chicago Booth’s Marianne Bertrand has conducted research on children’s changing attitudes towards gender roles that we wrote up in the Summer 2019 issue of Chicago Booth Review. The article was titled “Why Many Children Have More-Liberal Attitudes about Gender Roles.” It was written by CBR contributor Dwyer Gunn, and it’s read by Julie Granada-Honeycutt.
Reader: The traditional American family of the 1950s—characterized by a homemaker mother and a father employed outside the home—represents a shrinking percentage of US households. Almost 60 percent of married mothers in 2011 were employed outside the home, up from 25 percent in 1960—and almost a quarter of married mothers earned more than their husbands did, up from 4 percent, according to the Pew Research Center. Census data indicate nearly a quarter of children lived with only their mother in 2016, up from just 8 percent in 1960.
Many researchers have been looking at how this cultural shift, and the changing balance of economic power between men and women, has affected attitudes among adults. But what are the effects of this social shift on children? Research by Chicago Booth’s Marianne Bertrand suggests it’s leading many children to develop more-liberal attitudes toward gender roles.
Researchers have in recent years amassed evidence that the changing nature of the American family is causing tension in some households. For example, a study by Bertrand, Chicago Booth’s Emir Kamenica, and National University of Singapore’s Jessica Pan, a graduate of Booth’s PhD Program, suggests that US women who earn more money than their husbands are less likely to report happy marriages and are more likely to divorce.
To examine how changing gender dynamics are affecting children’s notions of gender roles, Bertrand analyzed multigenerational data from the NLSY79, a long-running survey launched by the Bureau of Labor Statistics in 1979. The BLS polled a group of almost 13,000 youths (initially aged 15 to 22) annually between 1979 and 1994. Since then, it has continued to contact the participants biennially—plus the children of female respondents.
The survey has always asked its respondents to react to six statements about traditional gender roles. The statements include, “A woman’s place is in the home, not the office or shop.” And, “It is much better for everyone concerned if the man is the achiever outside the home and the woman takes care of the home and family.” Bertrand used reactions from the children of female NLSY79 respondents to construct an index measuring attitudes about gender roles, with higher numbers corresponding to more liberal attitudes, and 24 representing the highest possible index score.
She finds that children of nonmarried mothers exhibited more-liberal gender-role attitudes. They were, for example, less likely to agree that “a woman’s place is in the home” or that women are happier when they “stay at home and take care of their children.” Having a nonmarried mother was associated with a 0.6 increase in the gender-role index for both male and female children.
For the children of married, working mothers, the effect of having a mom who’s employed outside the home differed by gender. Girls’ gender-role attitudes didn’t appear to be meaningfully affected by having a working mom. For boys, however, having either a married, working mother or a mother who was the primary breadwinner did meaningfully move attitudes in a more liberal direction. Having a primary breadwinner mother, in fact, is reflected by an increase of 1.2 in the gender-role index—a magnitude that’s comparable to the average gap in attitudes between girls and boys.
Several other factors are also correlated with gender-role attitudes. For boys, growing up in a household with a higher and more stable income was associated with more-liberal gender attitudes. A mother’s education levels and her own gender-role attitudes seemed to have a particularly strong effect on her daughters’ views.
Bertrand also finds that the family-work arrangements children experience from age 6 to 15 are more strongly correlated with gender-role attitudes than the arrangements they’re exposed to earlier in life. This finding provides support for the theory that mothers’ role modeling (their participation in the workforce, for example), rather than unobserved variables such as fathers’ gender-role attitudes, drove the results. Role modeling is likely to have a stronger effect as children grow older and are more aware of their mothers’ work lives, writes Bertrand.
Boys of working, married mothers who were employed in the formal labor market from economic necessity rather than personal preference displayed less of the liberal shift in gender-role attitudes, the research finds.
Hal Weitzman: Harvard’s Claudia Goldin is a leading economist who is an expert, among many other things, on the economics of gender. We were delighted to publish an essay by her in our Spring 2019 issue that links the division of labor in the home to women’s economic progress. The piece was called, “Tackle Gender Inequality at Home and at Work,” and it’s read by Julie Granada-Honeycutt.
Reader: In the 1970s, college-educated women in the United States began, as a group, to define success in terms of having a career and a family. By the 2000s, they had entered high-paying, male-dominated fields in large numbers. They are having children at a rate not seen since the late baby boom years.
But women still earn less than men, which we term gender inequality. And they give more hours to their families than men do, which we might call couple inequity. This holds for same-sex couples too—there is, on average, a certain amount of couple inequity between all partners.
Let me emphasize that most of what I have to say has to do with college-educated women and their quest, historically, for identity, meaningful employment, and family. And for this group, the real reason that both gender inequality and couple inequity exist is that the price of equity is so high. The cost of temporal flexibility, or controlling one’s hours, is substantial—and has become even more substantial with rising inequality.
The gender earnings gap widens a lot with time for these women. It widens at the time of marriage and at various joyous events, such as having children. Claims that this gap is primarily due to labor market bias have reached a fever pitch. There is no question that a lot of bad behavior exists in our workplaces, and we should do everything we can to wipe it out, but this isn’t the main cause of gender earnings inequality. Rather, the cause is the same as that which produces couple inequity.
The problem is that many jobs pay far more on an hourly basis when the work is longer, on call, rush, evening, weekend, or unpredictable. The lack of controllability is the crucial part, and the time and energy spent interferes with family commitments. Among the majority of highly educated couples with careers and children, the woman is the professional who is on call at home, while the man is the professional who’s on call in the office. And in consequence, he earns more than she does.
Sometimes, this earnings difference is a lot, maybe $30,000. This gives rise to the gender gap, and it produces couple inequity. If the difference were smaller, or smaller as a percentage of their earnings, they might decide: “Let’s just purchase couple equity. Because family equity is pretty cheap, we’re both going to be on call at home.” But the difference is often much larger.
Some people say, “Oh, you can contract out.” But you can’t contract out everything. There has to be some residual claimant—someone who is taking charge. And if you contract everything out, why have the kids in the first place?
So one member of the couple works the flexible, less-remunerative job. The other one works the less-flexible, more-remunerative job. The problem isn’t that women don’t compete or bargain enough. It isn’t that managers are so biased. It’s that these less-flexible jobs simply pay a lot more. And soaring earnings inequality since the 1980s has meant that these jobs are paying even more per hour. The problem, in many ways, is a combination of how work is structured and the norms that we inherit.
A history of career versus family
For a woman who graduated from college around the turn of the 20th century, a degree gave her the ability to be independent and not to marry in a world in which wives were supposed to stay at home—and usually did. An educated woman who wanted a job, let alone a career, didn’t get married. Her choice was to have a career or have a family. Few in this group managed both.
Of women who were born in the latter part of the 19th century and graduated from college around 1900 to the end of World War I, about half didn’t have children, and 30 percent didn’t marry. Still, few in this group, which I call cohort one, worked.
Next came a transitional cohort, aspiring to have a job and then a family. And then came a third cohort, born between the 1920s and the early 1940s, going to college from around World War II to the mid-1960s. In this group, married women could have a family first, followed by a pretty good job when their kids were grown. Marriage bars—the prohibitions against married women working in certain occupations such as teaching or clerical work—ended sometime in the 1940s. In addition, part-time work, which was important for women having a fair number of children and going to work when their kids were still in school, expanded greatly in the 1950s. This group had a family and then a job. Only 8 percent in this group never married over their lifetimes, and about 17 percent never had children.
Sometime around 1970, the pill allowed single women to delay marriage and to invest more in their careers before having kids—to aim for a career and then a family. The marriage age soared for this cohort, born between 1944 and 1957, as did the fraction of women with high-powered degrees and occupations. This is my cohort, graduating from college in the 1960s to the late 1970s. Many of these women put off family for too long. Although just 10 percent never married, many never had children. The childless rate for all women in this group peaked around 28 percent.
For the most recent cohort, the goal is not just family, not just work, but both together. This is the career-and-family group for whom the birth rate has increased.
The 100-year transition from career or family to career and family is bookended with two women, one who served in Congress and the other who is currently serving there. Jeannette Rankin was the first woman elected to federal office. Typical of those in her cohort who chose a career, she never married or had children. She had only her career, and it was an amazing one at that. She was elected in 1916 as a Republican from her state of Montana, just in time to vote against US entry into WWI. Rankin was also in Congress at the moment when she could vote for the right for women to vote. After we entered the war, she was not reelected, but she was elected again in 1940, in time to be the only vote against entry into WWII.
Senator Tammy Duckworth (Democrat of Illinois), the first senator to have a baby while in office and the first to bring a baby into an active session of Congress, is the other bookend.
I’ve computed the fraction of women in each cohort who achieved career and family. (I define career as having a job in which a person exceeds the 25th percentile of the male distribution of earnings given age and education, and achieves this for three years out of every five. For these purposes, I define family as having at least one child.) My conclusion is that, from the third to fifth cohorts, there’s been considerable improvement. In these three cohorts, women are experiencing more career and family over their life cycles.
But by and large, the achievement of career and family even in the fifth cohort doesn’t exceed 30 percent, about half of the level for the corresponding male cohort. The frustration of women in cohort five is that they’re still not reaching career and family success to a great extent. They’re doing somewhat better later on in their lives, but young women see that they have a battle, causing the recent explosion of concern.
The root of the earnings gap
Studies of various occupations suggest that the cost of flexibility is particularly substantial in some careers. In the case of MBA-degree holders, women are taking off somewhat more time than men are. Although women make far less than men in the financial and corporate sectors, most of the difference is due to their hours of work and, to some extent, to the amount of time they take off. When we say that women are working shorter hours, it’s still about 45–50 hours weekly. They’re just not working the jobs that demand super long hours. And the penalty in these sectors from taking any time off is harsh. The ability of individuals to work part-time is limited. And half of the women who claim to be working part-time are self-employed.
The situation is somewhat similar for doctors. In the Community Tracking Survey data set of doctors, 60–70 percent of physicians are in pediatrics, family medicine, and internal medicine. Female MDs are disproportionately in lower on-call specialties and are therefore working fewer hours. Among younger physicians (meaning less than 45 years old), females work almost 10 fewer hours weekly than males in just about every specialty. And for physicians older than 45, the number of hours for men goes down and the number of hours for women goes up, cutting the difference in half. What’s interesting to consider is that when you’re younger, you can work fewer hours and remain a doctor. But when you’re older and working somewhat more, you’re less likely to be in a leadership position, such as the head of your department.
The point is that many occupations have severe penalties for shorter and more-flexible hours. They give large windfalls to those who work hours that are less controllable, and this creates real problems for couples. The cost of couple equity becomes enormously high.
Therefore, one chooses the more flexible job with lower hours while the other chooses the job with less flexibility, more hours, and higher income. You produce gender inequality and couple inequity.
How can we fix this?
Any solution has to involve lowering the cost of the amenity in question, temporal flexibility. The simplest way of doing this is to create a perfect substitute for yourself. Use technology to pass information to a perfect substitute with little loss of fidelity. Take pharmacists as an example. You may hand in a prescription to Jane, but when you return to the pharmacy, you collect medication from Joe. It doesn’t matter because both pharmacists have information about all the prescriptions you have ever had under your insurance.
We could also create effective teams of substitutes, as has been done in certain health professions in which the cost of the professional has reached such a high level that organizations want there to be groups. Pediatrics is a perfect example. Pediatricians have formed groups so that they don’t have to be on call all the time. Veterinarians and anesthesiologists have done the same thing.
Is this commodifying important professions? Perhaps. But some of the highest-paid, most-well-trained professionals operate this way. When you have surgery, you may talk to your surgeon for months before, but you meet the anesthesiologist responsible for keeping you alive seven minutes before you go under. The anesthesiologist is in a group practice that has contracted with the surgeon or hospital and is available for that surgery.
The long road to the future has to involve a reduction in the cost of flexibility and probably has to involve some restructuring of jobs. And this is not a zero-sum game. Gender equality is not just about women. Men being on call at work and women being on call at home is not equitable for couples. When the cost of flexibility is lowered, we’re going to have solved the problem and achieved both gender equality and couple equity, but it’s going to take men demanding greater temporal flexibility and more control over the hours they work.
When people leave jobs, they often express to companies what prompted them to resign, or they may, in advance of leaving, demand more income for what they’re being asked to do. This can spur companies to think about substitutes and other organizational changes. That’s the way companies will have an incentive to lower the cost of temporal flexibility. And that’s a way to lead us to the road to the future.