The gender pay gap is the difference between what men and women earn, usually measured as median hourly pay, annual earnings, or income over a career. Economists break down the causes by separating raw differences in pay from differences linked to occupation, hours worked, work experience, industry, parenthood, promotion patterns, bargaining, and discrimination. That distinction matters because the headline gap is real, but it does not all come from a single source. In labor economics, the central question is not whether a gap exists; it does in nearly every advanced economy. The harder question is why it persists after accounting for measurable factors, and what remains unexplained.
When I have worked through compensation data with employers and public datasets, the first lesson is always the same: averages hide structure. A company can show equal pay within a job grade while still producing a large firmwide gap because men dominate higher-paid roles. A national dataset can show narrowing differences for younger workers while the lifetime gap remains large because women are more likely to reduce hours or leave the workforce after childbirth. Economists therefore analyze the gender pay gap at several levels at once: within the same job, across occupations, across industries, and across the life cycle.
This topic matters because pay influences retirement savings, household wealth, debt capacity, and economic growth. Lower earnings today compound into lower Social Security benefits, smaller pensions, and less capital accumulation later. At the macro level, barriers that keep skilled workers out of productive roles reduce labor supply and misallocate talent. Policymakers, employers, and families all make decisions using these numbers, so sloppy interpretation creates bad policy. A useful gender pay gap analysis must define the metric, identify the relevant comparison group, and explain the mechanisms clearly enough that readers can distinguish sorting effects from unequal treatment.
Economists usually start with two headline measures. The unadjusted gap compares all working men and women and captures the overall earnings difference in the labor market. The adjusted gap compares workers with similar observable characteristics such as education, occupation, tenure, and location. Both measures are important. The unadjusted gap reflects the world people actually experience. The adjusted gap asks whether women and men with similar measurable profiles are paid differently. Neither one, by itself, tells the full story, which is why serious analysis breaks the gap into components rather than relying on a single statistic.
How economists measure the gender pay gap
Measurement decisions shape the result. Economists often prefer hourly pay because annual earnings combine wage rates with hours worked, overtime, and time out of the labor force. Median pay is also common because it is less distorted by a small number of very high earners. In the United States, the Bureau of Labor Statistics and Census Bureau provide standard inputs, while researchers also use panel surveys such as the Panel Study of Income Dynamics and administrative tax data. Internationally, the OECD, Eurostat, and the International Labour Organization publish comparable series, though definitions differ enough that direct cross-country comparisons require caution.
The standard decomposition begins by estimating how much of the observed gap is associated with differences in characteristics and how much remains after those characteristics are held constant. A well-known method is the Oaxaca-Blinder decomposition, which separates the pay gap into an explained portion and an unexplained portion. Explained does not mean fair, and unexplained does not automatically prove discrimination in a legal sense. The explained part can reflect structural barriers that channel women into lower-paid work. The unexplained part can include discrimination, but also unmeasured variables such as performance differences, specialty choices, or job amenities that the dataset does not capture.
Researchers also look beyond wages. Total compensation can include bonuses, commissions, stock awards, pensions, health insurance, and paid leave. In finance, technology, law, and executive leadership, variable pay can produce larger gaps than salary alone. Promotions matter too. A modest annual difference in raise rates or promotion timing creates a much larger lifetime earnings gap through compounding. That is why economists study hiring, performance evaluation, assignment to revenue-generating work, and exit rates, not just paychecks.
Human capital, hours, and occupational sorting
One major source of the gender pay gap is occupational sorting, meaning men and women are distributed unevenly across jobs and industries. Occupations dominated by men, such as engineering, certain skilled trades, and senior finance roles, often pay more than occupations dominated by women, such as caregiving, education support, and administrative work. Economists do not treat this as a neutral preference story and move on. They ask how social norms, educational pipelines, licensing rules, hiring networks, and workplace culture shape those choices long before a wage is observed.
Hours also matter. Many labor markets pay a premium for long, inflexible, or unpredictable hours. Claudia Goldin’s work has shown that in some occupations earnings are non-linear: workers who put in very long hours at specific times earn disproportionately more than those who work fewer or more flexible hours. If women are more likely to seek flexibility because of caregiving demands, even equally skilled workers can end up on different earnings paths. In practical terms, a pharmacist in a role with standardized handoffs may face a smaller penalty for flexibility than a corporate lawyer whose client relationships reward constant availability.
Education no longer explains the gap in the simple way many people assume. In many countries, women have matched or exceeded men in college completion. Yet field of study still influences pay. Degrees in computer science, engineering, and some business specialties tend to lead to higher starting salaries than degrees concentrated in lower-paid sectors. Experience and tenure matter as well, but economists now emphasize that experience is not just years since graduation. Interrupted careers, reduced hours, and slower progression to supervisory roles alter accumulated human capital even when formal education is identical.
| Cause economists analyze | What it means | Typical effect on the gap |
|---|---|---|
| Occupation and industry | Men and women work in different jobs and sectors | Explains a large share of the overall gap |
| Hours and flexibility | Long or unpredictable hours are paid at a premium | Widens earnings differences, especially in high-skill roles |
| Experience and tenure | Time in the workforce and continuous advancement | Affects raises, promotion timing, and lifetime pay |
| Parenthood | Children change labor supply and employer behavior | Creates the largest divergence over the life cycle |
| Discrimination | Unequal treatment in pay, hiring, or promotion | Often appears in the unexplained residual |
The motherhood penalty and the life-cycle gap
The sharpest divergence in earnings often appears around the first child. Economists call this the motherhood penalty: a decline in women’s earnings relative to men’s associated with childbirth and caregiving. Studies using administrative data from Denmark, the United States, Germany, and the United Kingdom consistently find that the gap expands after children arrive, not mainly at labor market entry. The mechanisms include reduced hours, career interruptions, movement into more flexible but lower-paid jobs, and employer assumptions about availability and commitment.
By contrast, men sometimes experience a fatherhood premium, especially in workplaces that interpret fatherhood as a signal of stability or commitment. This does not mean every father gets a raise because he has children. It means that, on average, institutional and cultural responses to parenthood differ by gender. In my experience reviewing pay structures, the biggest long-run effects often come from project allocation and promotion tracks during the years when children are young. Missing one stretch assignment, one sales territory, or one path to profit-and-loss responsibility can change earnings trajectories for decades.
Childcare costs and leave policy strongly influence these outcomes. Where childcare is scarce or expensive, the financial case for one parent, usually the mother, to scale back work becomes stronger. Where parental leave is available only to mothers, employers may statistically expect women of childbearing age to take longer absences, which can affect hiring and advancement decisions. Countries with well-designed, gender-neutral leave and strong childcare systems tend to show higher female labor force participation, though no policy fully erases the gap because norms inside households and workplaces still matter.
Discrimination, negotiation, and firm behavior
Discrimination enters the analysis in several ways. There can be direct pay discrimination, where equally productive workers receive different compensation. There can also be hiring discrimination, promotion discrimination, and evaluation bias. Audit studies, in which otherwise similar résumés are assigned different gender signals, have found unequal callback rates in some sectors. Performance review research has documented differences in how identical behavior is described, with men more often framed as decisive and women as abrasive. Economists treat these patterns as mechanisms that affect earnings even when base pay within a narrow role appears similar.
Negotiation is often mentioned, but the evidence is more nuanced than popular advice suggests. Men may negotiate more often in some settings, yet outcomes depend heavily on organizational rules. In transparent pay systems with tight bands, individual negotiation has less room to create gaps. In opaque systems with discretionary bonuses, initial offers and manager judgment matter more. Women can also face social penalties for negotiating assertively. That is why economists focus on institutional design. Standardized salary bands, posted ranges, structured interviews, and calibrated promotion reviews reduce the degree to which unequal bargaining power turns into unequal pay.
Firm effects are especially important. Recent research shows that some employers systematically pay more than others for similar workers. If men are more likely to work at high-paying firms and women are more concentrated in lower-paying firms, the overall gap widens even without unequal pay inside each firm. This sorting can happen through recruiting networks, geographic constraints, scheduling practices, or workplace cultures that are less compatible with caregiving. From a policy perspective, improving access to high-wage firms can matter as much as policing pay equality within firms.
What the unexplained gap really means
Readers often ask whether the unexplained portion is “the real” gender pay gap. The better answer is that it is the residual left after controlling for the variables in a given model. That residual is informative, but it is not pure discrimination and it is not trivial noise. If a dataset lacks detailed information on exact tasks, client portfolios, career interruptions, or willingness to relocate, those omitted factors will sit in the unexplained component. At the same time, if measured variables themselves reflect earlier discrimination, then controlling for them can understate the role of unequal treatment.
For example, if women are less likely to enter a high-paid specialty because mentors, hiring managers, or training pathways discouraged them earlier, an adjusted model that controls for specialty choice may label that portion explained. Economically, however, the process that created the specialty gap may still involve bias or constraint. This is why careful researchers pair statistical decompositions with institutional analysis. They ask what decisions happen upstream, who controls them, and how incentives are structured. Numbers identify patterns; context explains their origin.
Another reason the unexplained gap varies across studies is that methods differ. Some analyses compare all full-time workers. Others narrow the sample to the same occupation, firm, or job level. Some use annual pay; others use hourly wages. Results can all be accurate within their frame. The mistake is treating one estimate as the single truth. A robust interpretation looks for convergence across methods: where the gap appears, when it widens, and which mechanisms repeatedly account for the change.
What reduces the gender pay gap
Effective responses target structure, not slogans. Pay transparency laws help by making hidden disparities easier to spot and by constraining arbitrary offers, though transparency alone does not solve occupational segregation or caregiving pressures. Childcare availability, predictable scheduling, and parental leave that encourages fathers to take time off reduce the motherhood penalty more directly. In firms, structured compensation systems, formal promotion criteria, and regular pay equity audits are among the most evidence-based interventions. When I have seen organizations make real progress, they did not rely on one annual analysis. They tracked hiring slates, starting pay, assignments, performance scores, promotions, and exits in a single dashboard.
Education and pipeline programs matter too, especially in high-growth, high-wage fields. But the goal is not simply to tell women to choose different majors. It is to remove barriers in training, internships, sponsorship, and early-career advancement so that choices are genuinely open. Flexible work can help retain talent, yet it must be designed carefully. If remote or reduced-hour tracks become stigmatized and disconnected from advancement, they preserve the gap under a more convenient label. The strongest results come when flexibility is normalized for everyone and output is measured clearly.
The clearest takeaway is that the gender pay gap is not one problem with one cause. Economists break it down into hours, occupations, parenthood, firm sorting, promotion patterns, and discrimination because each mechanism points to a different remedy. If you want to understand the numbers in any article or workplace report, ask three questions: what measure is being used, what factors are being controlled for, and what decisions occur before pay is set. Use that framework to read related economics coverage, compare studies carefully, and challenge simple explanations with better evidence.
Frequently Asked Questions
What is the gender pay gap, and how do economists usually measure it?
The gender pay gap is the difference between what men and women earn, but economists are careful about how they define that difference. The most common starting point is a “raw” or “headline” gap, which compares average or median earnings across all working men and women without first adjusting for other factors. This can be measured in several ways, including median hourly pay, annual earnings, weekly wages, or even total income over a lifetime. Each measure tells a slightly different story. Hourly pay can show what workers earn for a given hour of labor, while annual earnings also reflect differences in hours worked, weeks worked, time out of the labor force, and part-time versus full-time employment.
Economists often prefer median measures because they are less distorted by a small number of very high earners. They also look closely at hourly wages when the goal is to compare pay rates more directly. At the same time, annual and career earnings matter because they capture the broader economic reality people experience. A worker who earns the same hourly rate but works fewer hours or experiences more interruptions in employment can still end up with substantially lower yearly or lifetime income.
In labor economics, the key point is that the headline gap is real, but it is not automatically explained by one single cause. A raw pay gap can reflect differences in occupation, industry, experience, schedule flexibility, family-related career interruptions, access to promotions, and discrimination. That is why economists do not stop at the first number. They use the raw gap as a starting point, then break it down to understand how much is associated with measurable job-related factors and how much remains unexplained after those factors are taken into account.
Why do economists separate the “raw” gender pay gap from the “adjusted” pay gap?
Economists separate the raw and adjusted gender pay gaps because the two measures answer different questions. The raw gap describes the overall difference in earnings between men and women in the labor market. It is useful because it reflects the actual outcomes people experience in real life. If women, on average, bring home less income than men, the raw gap captures that fact. However, it does not tell us why the difference exists.
The adjusted gap goes one step further. It uses statistical methods to compare men and women who are similar in specific observable ways, such as education, work experience, occupation, industry, region, job tenure, and hours worked. This helps economists estimate how much of the overall gap is associated with differences in where people work or how their careers unfold, rather than simply differences in pay for similar work. In other words, the adjusted gap is an attempt to compare more like with like.
This distinction matters because some public discussions treat the entire headline gap as if it comes only from unequal pay for the same job, while others dismiss the gap entirely by pointing to occupational or lifestyle differences. Economists generally reject both extremes. If women are concentrated in lower-paying occupations, if they are more likely to reduce hours after becoming parents, or if they face barriers to advancement, those patterns are themselves important parts of the economic story. The adjusted gap is narrower by design, but it does not make the rest of the raw gap irrelevant. Instead, it helps identify which portion is linked to measurable characteristics and which portion remains unexplained and may reflect harder-to-measure factors, including discrimination, workplace norms, or structural constraints.
What are the main factors economists look at when breaking down the causes of the gender pay gap?
When economists analyze the gender pay gap, they usually examine a cluster of related factors rather than searching for one universal explanation. Occupation is one of the biggest. Men and women are often distributed unevenly across fields, and some occupations pay much more than others. Industry matters too, because the same type of worker may earn different pay in finance, manufacturing, education, healthcare, or retail. Hours worked are another major factor, especially when the comparison uses annual earnings instead of hourly wages. Workers who put in more overtime, accept irregular schedules, or remain continuously employed throughout the year may earn more overall even if hourly rates are similar.
Experience and job tenure are also central. Time spent out of the labor force, part-time work, and interrupted career paths can affect wage growth over time. Parenthood is especially important in this area. Economists often find a “motherhood penalty,” meaning women’s earnings tend to slow or decline after having children, while men may experience a smaller effect or even a “fatherhood premium” in some settings. Promotion patterns matter as well. If women are less likely to move into senior, higher-paid roles, wage differences widen over the course of a career.
Researchers also study bargaining, workplace flexibility, and compensation structure. In some jobs, earnings rise sharply for workers who can work long, inflexible, or unpredictable hours. If caregiving responsibilities fall more heavily on women, those jobs may reward availability in ways that indirectly widen the gap. Finally, economists examine discrimination, both direct and indirect. Direct discrimination can involve paying women less for comparable work or evaluating them differently. Indirect discrimination can arise through biased hiring, promotion standards, task assignments, or workplace cultures that make advancement harder. The overall pay gap usually reflects a combination of these influences interacting over time, not a single isolated cause.
Does an “unexplained” gender pay gap prove discrimination?
Not automatically, but it is a serious signal that economists take seriously. The unexplained portion of the gender pay gap is the part that remains after adjusting for measurable characteristics such as education, occupation, experience, hours, industry, and location. Because the available data never capture every relevant factor, economists are careful not to say that every unexplained difference is definitive proof of discrimination. Some of it may reflect variables that are difficult to observe well, such as exact job duties, career preferences, informal mentoring, workplace culture, or detailed measures of productivity.
That said, the unexplained gap should not be dismissed. If men and women with similar observable qualifications still earn different amounts, economists ask why. In many cases, discrimination is one plausible explanation. It can appear in hiring decisions, promotion opportunities, performance reviews, salary negotiations, bonus structures, and assumptions about commitment or availability. It may also operate subtly. For example, if women are steered away from high-visibility assignments or are penalized more heavily for negotiating, those patterns can depress earnings without appearing as a simple line item in a dataset.
Economists often combine pay-gap analysis with evidence from audit studies, field experiments, promotion data, and employer-level research to get a fuller picture. When several kinds of evidence point in the same direction, the case for discrimination becomes stronger. So the best way to understand the unexplained gap is not as a courtroom verdict on its own, but as an important indicator. It tells researchers that measurable differences do not account for the entire outcome and that deeper institutional, behavioral, or discriminatory mechanisms may still be at work.
Why does the gender pay gap often widen over the course of a career?
The gender pay gap often starts relatively small early in working life and then widens over time because pay differences accumulate through career progression, family transitions, and promotion systems. Early on, men and women may have similar education levels and may begin in comparable entry-level roles. But as careers develop, small differences in raises, assignments, hours, and advancement opportunities can compound. A worker who gets one more promotion, one more bonus opportunity, or a few more years of uninterrupted full-time work can pull ahead substantially over a decade or more.
Parenthood is one of the most important reasons the gap expands. Many economists find that women’s earnings trajectories often change after children arrive. Some reduce hours, shift into more flexible roles, decline travel-heavy positions, or take career breaks for caregiving. Even when these decisions are rational responses to family needs, they can slow wage growth and reduce promotion prospects. Meanwhile, workplaces may reward long, continuous, inflexible hours, which tends to favor workers with fewer caregiving constraints. If men are less likely to face those constraints, or are perceived as more available, they may progress faster into higher-paying positions.
The widening gap also reflects occupational sorting within organizations. Men and women may start in the same company but end up on different ladders, with men more likely to move into leadership or revenue-generating roles that command higher pay. Differences in sponsorship, networking, evaluation standards, and access to stretch assignments can all matter. Over time, these forces create a cumulative effect: the gap is not just about a single paycheck but about how opportunities, responsibilities, and rewards are distributed across an entire career. That is why economists often emphasize career dynamics rather than only focusing on pay at one moment in time.