The gender pay gap is the difference between what men and women typically earn, and in the United States it exists because of a mix of factors — different occupations and hours, the effect of having children, and a residual portion that economists cannot fully explain and often attribute in part to discrimination. In 2024, the median woman working full time in the U.S. earned about 85 cents for every dollar earned by the median man, according to Pew Research Center, a gap that has narrowed only slightly since 2003, when it stood at 81 cents. That headline figure is cited constantly, in campaign speeches and corporate diversity reports alike, but economists who study it closely have found the underlying story is more complicated, and more genuinely debated, than one number suggests.
The raw number is not the same as the adjusted number
The 85-cents figure is what researchers call the “unadjusted” gap: a simple comparison of median earnings across all men and all women, with nothing else controlled for. Economists also calculate an “adjusted” gap that statistically accounts for measurable differences such as occupation, industry, education, and hours worked. Once those factors are held constant, most academic estimates put women’s earnings closer to 95 to 98 cents on the male dollar. The gap between the raw and adjusted numbers matters because it shows that a large share of the headline gap is explained by structural differences in who works where, not by identical workers being paid differently for identical jobs — though a meaningful, long-debated portion still is unexplained by any observable characteristic economists have found a way to measure.
Different jobs and different hours explain a large share
A significant chunk of the raw gap traces to the fact that men and women, in aggregate, still work in different jobs and different schedules. Women remain overrepresented in lower-paying fields such as education, retail, and health support roles relative to their share of the overall workforce, while men are more concentrated in higher-paying fields like engineering and finance. The Bureau of Labor Statistics has also found that, on days worked, employed men log somewhat more hours on average than employed women, partly reflecting different rates of part-time work. Research on “occupational segregation” adds a further wrinkle: jobs performed predominantly by women tend to pay less even after controlling for the skill and education those jobs require, a pattern some economists attribute to how the labor market has historically valued that kind of work. In other words, the gap is partly about which doors different workers walk through, and partly about what happens to pay once women are already behind one of them.
Parenthood is where the gap widens most sharply
A substantial body of research points to parenthood as a pivot point. Sociologist Shelley Correll’s widely cited audit studies found that mothers received fewer callbacks and lower salary offers than childless women with identical resumes, while fathers were sometimes offered more than childless men — a pattern researchers describe as a “motherhood penalty” paired with a “fatherhood bonus.” Census Bureau linked survey data similarly show earnings for women often diverge most sharply from men’s in the years after a first child, coinciding with reduced hours, career interruptions, or a shift toward more flexible but lower-paying roles. The consequence is that the gender pay gap is not static across a career: it tends to be narrow early on and widen after parenthood, which is part of why average, career-long figures look larger than gaps measured among workers just starting out.
Negotiation, bias, and the part nobody can fully explain
Other researchers point to more subtle dynamics. A frequently cited Carnegie Mellon study found men initiated salary negotiations considerably more often than women, though follow-up research also found that women who do negotiate can face social penalties men do not, which helps explain why the gap persists even where negotiation is technically open to everyone. Separately, audit studies of hiring and performance evaluation — including work finding that identical resumes are rated differently depending on the name attached — are often cited as evidence of ongoing bias. Economists Francine Blau and Lawrence Kahn, whose long-running research on the topic is among the most cited in the field, have estimated that a meaningful share of the pay gap, on the order of a third or more by some counts, is not explained by any factor economists have found a way to measure. That residual is the crux of the disagreement: some attribute it to discrimination, others to unmeasured differences in job preferences, negotiation style, or career interruptions that statistics simply cannot capture.
Why the debate persists
The gap has multiple contributing causes that are difficult to fully separate statistically, and there is legitimate disagreement among economists about how much of it reflects discrimination versus choice, and how much of “choice” itself is shaped by social expectations around who takes time off, who negotiates, and which jobs are considered women’s work in the first place. That is why the honest answer to “why does the gender pay gap exist” is not a single villain or a single fix, but a layered set of overlapping causes, some well quantified and some still argued over.