The average difference between the remuneration for men and women who are working 35+ hours per week year round. The median salary for women is roughly 22 percent lower than the median salary for men into 2018. The data cited in the gender pay gap looks only at the median earnings of full-time wage and salaried workers. It doesn’t differentiate really important factors, such as education, occupation, experience, and hours, which account for nearly all of the differential in earnings between men and women. Women are more likely to take less pay for more flexibility and men are more likely to work overtime hours for higher pay. It turns out that accounting for all these factors eliminates nearly all of the gender pay gap. Politicians continue to repeat the 22% gap either out of ignorance or as propaganda meant as another divisive tool of the left.
I’m pretty sure you’ve heard this kind of statistic before. It’s been repeated thousands, if not millions, of times.
POLITICAL AD: The gender pay gap is real and women still earn about 77 cents for every dollar a man earns for working the same job.
BARACK OBAMA: Today, women make up about half our workforce. But they still make 77 cents for every dollar a man earns. That is wrong. And in 2014, it’s an embarrassment. Women deserve equal pay for equal work.
In 1990, Claudia Goldin became the first woman to get tenure in the Harvard economics department. In 2014, she served as president of the American Economic Association, the AEA. Her lecture at the annual AEA meeting was called “A Grand Gender Convergence: Its Last Chapter.” It’s deceptively easy to calculate how much—or how little—women in the United States earn relative to men. “You take everyone who’s working 35 or more hours a week for the full year, find the median for women, find the median for men, and divide,” says Goldin, explaining how to arrive at the ratio repeated by public officials: 78 cents to the dollar. “It’s very simple.”
“It answers a particular question,” she says, “but it doesn’t say that men and women are doing the same thing. It doesn’t say that they’re working the same amount of time, the same hours during the day, or the same days of the week.” The rhetoric of politicians, and policy prescriptions meant to close the gender wage gap, assume that pay disparities are created primarily by outright discrimination by employers, or by women’s lack of negotiation skills. Goldin has a less popular idea: that the pay gap arises not because men and women are paid differently for the same work, but because the labor market incentivizes them to work differently.
Data is also subject to human error. Comparisons between survey data and administrative records reveal substantially underreporting of income within some of the most widely used survey data. Consequently, the data disregards substantial changes, such as large gains in women’s retirement incomes. And finally, data isn’t the supreme indicator, because not everything comes with a price tag or pay stub. What is the value of a flexible work schedule; a job with huge upward-mobility potential; particular benefits packages; the ability to tap into flexible, sharing-economy labor platforms, such as Uber and Airbnb; or to access new business platforms, such as Etsy for additional income? Workers who seek these job characteristics often do so despite lower pay. But those intentional choices don’t show up in the statistics.
If a woman has the exact same job title as a man, but works 30 hours a week instead of 40, and sets her own hours and telecommutes, her paycheck likely won’t match that of the man’s—nor should it. One of the job qualities that women—particularly mothers—value most is flexibility. Flexibility is a difficult job feature to measure, but that’s exactly what a group of economists recently did using data from the Uber ride-hailing company. After analyzing data from more than 1 million registered Uber drivers, the authors tagged the average value of being able to set one’s own work schedule on an hour-by-hour and minute-by-minute basis at $150 per week. That’s the equivalent of $7,800 per year, or almost 20 percent of the median earnings of women in the U.S.
In essence, this is the value of choice. It’s not the same value for everyone, but it shows that many workers are willing to sacrifice a lot in terms of pay for more flexibility and choice. On the opposite side, some employers are willing to pay a high price for flexibility from their employees—to log long hours and to work day or night. Economist Claudia Goldin has found evidence of “part-time penalties” in certain very high-income fields. This happens when certain companies—those in finance and law, for example—pay employees who work 80 hours a week more than twice as much as they pay those who work 40 hours per week. This likely has to do with certain employers’ need for employees to respond at all hours or to log double or triple time when needed, coupled with employees’ demand for higher pay when sacrificing so much of their own time and flexibility.
Anecdotal evidence and the choices women and men make suggest that women value job choices more than men and that their preference for greater flexibility accounts for some—if not all—of the remaining pay gap between men and women. Pay regimes based on factors such as job titles or “equivalent work” would take away businesses’ freedom to determine the value of their work and undo decades of women’s progress by imposing one-size-fits-all jobs that take away women’s—and all workers’—freedom to negotiate pay in exchange for personal priorities.
Consider a couple graduating together from a prestigious law school, and taking highly paid jobs at firms that demand long hours. The evidence suggests they’re likely to begin at similar salaries. But a few years later, Goldin says, one of them—more likely the woman—may decide to leave for a smaller practice with fewer hours and more flexibility in scheduling. In that new job, research suggests, she’s likely to earn less per hour than her partner. Goldin calls this phenomenon non-linearity, or a part-time penalty: the part-timer works half the time her partner does, but earns less than half his salary.
It isn’t clear, she says, why firms compensate on a non-linear scale in the first place. “Why would anyone pay for that?” she asks. Apart from scenarios in which a client might want a lawyer available at all hours, day or night—during a merger or acquisition, say—and must offer a hefty premium for that unrestricted access, she says, “It’s a question I don’t have a particularly good answer to.”
Non-linear compensation prevails in the corporate sector, finance, and law, where employees are incentivized to work double or triple a traditional full-time schedule, because their time is better compensated per hour when they work longer hours. That compensation structure makes it more lucrative for one partner to work 80 hours and the other not to work at all than for both of them to work 40 hours each. If both partners opt for 40-hour weeks so they can share responsibilities at home, Goldin says, “lots of money is going to be left on the table,” which is why she believes so many couples don’t.
Non-linearity helps explain why most of the gender pay gap occurs within professions, Goldin adds. The distribution of men and women in different occupations accounts for only 15 percent of the gap, and the remaining 85 percent arises within occupations. (For college graduates, those numbers are 35 percent and 65 percent, respectively.) In science and health professions, though, workers are more likely to be compensated at a constant rate for additional time worked, and the ratio of women’s earnings to men’s is higher—about .892. For occupations in business and finance, the ratio is .787, and for lawyers, .815, closer to the national gender wage gap.
Improvements in technology have made it easier for some health and science professions to substitute workers for one another in a single job, which reduces the cost to companies of offering a flexible-hours option to employees. Goldin calls pharmacy “the most egalitarian profession” because it shows nearly perfectly linear compensation and one of the smallest gender pay gaps of any field. “Pharmacy has no part-time penalty,” she says. Structural changes, such as centralized computer records and standardization of drugs, allow one pharmacist to take over easily for another without compromising the quality of work. And because it’s easy for pharmacists to work part-time, women are less likely to have to leave their jobs to care for their families, a decision that can make it difficult to reenter the workforce later.
Goldin believes other fields could narrow their gender wage gaps, too, if they did not have an incentive to pay workers disproportionately more for working more. How to induce change in the labor market isn’t obvious. Why can’t you convince clients, she asks, that your employees are like puzzle pieces, each knowing everything the others know, so they’re good substitutes for each other? “As their labor costs mount,” she suggests, firms “will figure out how to make workers better substitutes for each other,” Technological change might also play a role, doing for law, perhaps, what it’s done for health professions, and making it easier for lawyers to hand off clients to one another. But in some cases, Goldin concedes, it may not be possible to embrace this modular model: “We don’t want the president of the United States to be a part-time president.”
As for policy interventions to close the gender earnings gap—a California law makes it illegal to retaliate against employees for sharing information about their pay, for example—“That’s probably a good thing,” Goldin says. “If the fruit is low-hanging, by all means pick it.” But she balks at the suggestion that regulation can fix what she sees as a labor-demand problem. Creating an egalitarian workplace, she believes, will depend primarily on reducing the cost of offering time flexibility to workers—securing equal pay for equal work, in the strictest sense.
The big question of the gender pay gap has to be broken down into a set of smaller questions. And then you have to find the data to answer them. When someone like Claudia Goldin does that, it’s pretty obvious that the statistic cited by everyone from Sarah Silverman to President Obama isn’t quite right. Because women aren’t getting paid twenty-some percent less than men for doing the same work. They are, however, often doing different work, or work that affords more flexibility — which tends to pay less.
Now, it may be that if you put a dollar value on the flexibility, it could offset a lot of the actual, salary dollars. In any case, there would seem to be kinda-sorta good news here, which is that discrimination doesn’t seem to be the main culprit in the gender pay gap. Or at least it’s hard to find a smoking gun, as Claudia Goldin says. But not impossible.
A hack of Sony Pictures e-mails showed that the actresses Jennifer Lawrence and Amy Adams got fewer back-end points than their male counterparts in the film American Hustle. Interestingly, when Lawrence later wrote about this revelation, she largely blamed herself. “I failed as a negotiator because I gave up early,” she wrote. “I didn’t want to keep fighting over millions of dollars that, frankly, due to two franchises, I don’t need… But if I’m honest with myself, I would be lying if I didn’t say there was an element of wanting to be liked that influenced my decision to close the deal without a real fight. I didn’t want to seem ‘difficult’ or ‘spoiled.’”
So, that’s another component to consider when you’re talking about the gender pay gap — that even in the absence of outright discrimination, the playing fields are not necessarily equal. Some of the most compelling evidence for this fact comes from the late 1990s, a study that Claudia Goldin did with Ceci Rouse. The paper was called “Orchestrating Impartiality.”