What Is the Gender Pay Gap: What the Data Actually Shows

Published 2026-04-14 · Updated 2026-04-15

✓ Fact-checked by the Kalenux editorial team

The gender pay gap is the difference in average or median earnings between men and women in the workforce -- and in the United States in 2024, women working full-time year-round earned approximately 83 cents for every dollar earned by their male counterparts, according to the U.S. Census Bureau. Few topics in labor economics generate as much heat and as little light. On one side, critics argue that the "77 cents on the dollar" figure is misleading -- a statistical artifact that disappears once you control for occupation and hours worked. On the other side, defenders argue that controlling for those variables is itself problematic, since occupation choice and hours worked are themselves shaped by gender inequality. Both sides have a point. Neither is complete.

The truth, as documented by decades of careful labor economics research -- most notably by Harvard economist Claudia Goldin, who won the 2023 Nobel Prize in Economic Sciences for her work on women's labor market outcomes -- is more nuanced than either camp acknowledges. There is a real gender pay gap. It has multiple causes operating at different levels. Different measures capture different dimensions of the problem. And understanding what the data actually shows is more useful than picking a side in a political argument.

This article examines the evidence: the distinction between raw and adjusted gaps, the groundbreaking research that earned Goldin the Nobel Prize, the mechanisms of occupational segregation and the motherhood penalty, what discrimination does and does not explain, how countries that have reduced the gap did so, and what the most effective interventions look like.

"The gender earnings gap is not a simple story of equal pay for equal work, nor is it a simple story of women freely choosing lower-paid paths. It is a story about how labor markets, career structures, and social institutions interact with the distribution of unpaid caregiving." -- Claudia Goldin, Career and Family: Women's Century-Long Journey Toward Equity (2021)

The Raw Gap vs. the Adjusted Gap: Two Different Questions

The most commonly cited figure in the United States -- that women earn roughly 82-84 cents for every dollar men earn -- is the raw (unadjusted) gap. It is calculated by comparing the median annual earnings of full-time, year-round female workers to the median annual earnings of full-time, year-round male workers, regardless of what those workers do, how many hours they work within "full-time," or their level of experience and education.

The U.S. Census Bureau's Current Population Survey has tracked this figure for decades. In 1960, the raw gap was approximately 40 cents on the dollar. By 1990, it had narrowed to roughly 28 cents. By 2023, it stood at approximately 16-18 cents. The trend is clearly improving, but the pace of improvement has slowed dramatically since the 1990s -- what researchers call the "stalling" of convergence.

The adjusted gap attempts to control for factors that explain some of the raw difference:

Occupation and industry
Hours worked per week (within full-time employment)
Years of experience and job tenure
Educational field (not just level of education)
Employer size, sector, and geographic location

When these controls are applied, the gap shrinks substantially -- to somewhere between 2 and 8 cents on the dollar in most well-designed studies, with the range depending on exactly what is controlled for and how. The Glassdoor Economic Research team (Chamberlain, 2016) found a controlled gap of about 5.4 cents after controlling for occupation, industry, city, experience, and job title. PayScale's 2024 analysis found a controlled gap of approximately 1 cent for workers in identical roles at identical companies -- though this extremely narrow comparison answers an extremely narrow question.

Measurement	Approximate U.S. Gap	What It Captures
Raw annual earnings gap	~16-18 cents	Total earnings difference including all causes
Raw weekly earnings gap	~14-16 cents	Reduces some of the hours-worked difference
Occupation-controlled gap	~10-12 cents	Removes between-occupation differences
Fully adjusted gap	~2-8 cents	Residual within identical jobs with similar experience

Why Both Measures Matter

The political debate often proceeds as if one of these measures is "the real" pay gap and the others are either exaggerations or excuses. This framing is mistaken. Both the raw and adjusted gaps measure real phenomena; they simply answer different questions.

The raw gap tells you how much less money women have in their bank accounts at the end of the year. It captures the cumulative effect of every factor -- occupational choice, hours worked, career interruptions, discrimination, negotiation patterns -- that contributes to earnings differences. For understanding women's economic position in society, the raw gap is the relevant measure.

The adjusted gap tells you how much of that difference persists within comparable job categories after accounting for measurable factors. For understanding whether employers are paying women less for the same work, the adjusted gap is more informative -- though it cannot answer that question definitively, because some relevant factors are unmeasurable.

The critical insight, emphasized by labor economists including Francine Blau and Lawrence Kahn of Cornell University in their comprehensive 2017 review in the Journal of Economic Literature, is that the factors being "controlled for" in adjusted analyses are themselves partly products of gender inequality. If women are steered away from high-paying occupations by socialization, workplace culture, or hiring discrimination, then controlling for occupation removes part of the inequality from the analysis rather than explaining it away.

Claudia Goldin's Nobel Research: The "Greedy Jobs" Problem

The most important recent advance in understanding the gender pay gap came from Harvard economist Claudia Goldin, whose decades of research earned her the 2023 Nobel Prize in Economic Sciences -- only the third woman to receive the economics Nobel and the first to receive it solo.

The Long Arc of Women's Labor Force Participation

Goldin's historical research, spanning her 1990 book Understanding the Gender Gap through her 2021 Career and Family, traced American women's labor force participation and earnings from the late 18th century to the present. She identified five distinct phases of women's economic participation, each characterized by different constraints and different patterns of progress:

Late 1700s-early 1900s: Women worked primarily in household production and manufacturing; participation declined as household income rose.
1900-1950: The "quiet revolution" began; women entered clerical and service work, but most left the labor force upon marriage.
1950-1970: Married women entered the labor force in large numbers, but primarily in secondary-earner roles.
1970-1990: Women began pursuing careers rather than jobs, entering professions (law, medicine, business) and investing in education with long-term career plans.
1990-present: Women achieved near-parity in education and entry-level hiring, but a persistent earnings gap emerged in the years after college graduation.

The Last Chapter Problem

The most important finding for the contemporary pay gap is what Goldin calls the "last chapter" problem: even as women have achieved near-parity in educational attainment (women now earn approximately 58% of bachelor's degrees and the majority of advanced degrees in the U.S.), the earnings gap persists and actually widens in the years after college graduation -- precisely when workers in their 30s are starting families.

Goldin's analysis of earnings data from elite MBA graduates (published with Marianne Bertrand and Lawrence Katz in the Quarterly Journal of Economics, 2010) found that male and female MBA graduates earned nearly identical salaries immediately after graduation. Ten years later, the women earned 60% of what the men earned. The divergence was almost entirely explained by two factors: career interruptions (even brief ones) and reduced hours -- both strongly correlated with childbearing.

Greedy Jobs and Nonlinear Pay

The mechanism Goldin identifies is what she calls "greedy jobs": positions that pay disproportionately -- nonlinearly -- for long hours, continuous availability, and temporal inflexibility. In corporate law, finance, consulting, and corporate management, an employee who works 60 hours per week does not earn modestly more than one who works 40 hours per week. They earn vastly more, because the compensation structure rewards the marginal hour of availability far above its direct productivity value.

The premium is for the willingness to be always on call, to take the 7 AM call, to stay late without notice, to be available on weekends. These positions are not inherently incompatible with women's preferences or abilities. But they are structurally incompatible with primary caregiving -- the role that continues to fall disproportionately on women in households with children. A parent who needs to be home by 6 PM on most evenings cannot compete for the greedy job premium, regardless of how talented they are or how productive their 40-50 hours are.

Goldin identifies pharmacy as a sector that has dramatically reduced the within-occupation gender pay gap by shifting toward standardized, output-based work that does not penalize flexible scheduling. A pharmacist who works 20 hours per week earns proportionally the same per hour as one who works 50 hours, because the work is structured around fungible tasks rather than client relationships and availability premiums. Corporate law and finance remain sectors with large within-occupation gaps, precisely because their economic logic rewards temporal inflexibility.

Occupational Segregation: The Pay Gap Before Workers Are Hired

A substantial share of the raw pay gap operates not within occupations but between them. Men and women remain highly segregated across occupations and industries, and female-dominated occupations pay systematically less than male-dominated ones requiring comparable education, skill, and responsibility.

Paula England's Research on Devaluation

Sociologist Paula England of New York University has studied occupational segregation and pay for over three decades. Her research, published in journals including the American Sociological Review and Social Forces, finds that the wage penalty for working in a female-dominated occupation persists after controlling for working conditions, physical demands, human capital requirements, and numerous other factors. The penalty is not entirely explained by skill differences; a significant portion reflects historical social processes by which jobs done predominantly by women came to be valued less.

A striking feature of England's research is what she calls the "devaluation of feminized work": as women entered fields that were previously male-dominated -- biology, veterinary medicine, law -- those fields' relative pay tended to decrease. As men entered fields that were previously female-dominated -- nursing, teaching in certain contexts -- relative pay increased. The pattern, documented in a 2009 study by Levanon, England, and Allison in Social Forces, suggests that part of what determines an occupation's pay is who does it, not just what it requires.

The Persistence of Segregation

Occupational segregation in the U.S. has declined since the 1970s but has not disappeared. The Duncan Index of Dissimilarity -- a measure of how many workers would need to change occupations for men and women to be identically distributed across jobs -- has fallen from approximately 68 in 1970 to approximately 51 in 2020, according to data from the Institute for Women's Policy Research. Progress has been real but incomplete.

Women are overrepresented in:

Healthcare support occupations (89% female)
Education, particularly K-12 teaching (74% female)
Administrative and clerical work (72% female)
Social work and human services (81% female)

Women are underrepresented in:

Engineering and computer science (approximately 20-25% female)
Executive and senior management (approximately 29% female at Fortune 500 companies, per Catalyst 2024 data)
Skilled trades and construction (approximately 4% female)
Finance and investment management (approximately 30% female)

The causes of this segregation are themselves debated. They include early socialization and gender norms, differential recruitment and hiring practices, workplace cultures that are unwelcoming to minority-gender workers, and genuine differences in average preferences -- though the size and origins of these preference differences are contested, and cross-national variation suggests that social context plays a larger role than innate preference.

The Motherhood Penalty and the Fatherhood Bonus

One of the most robust findings in gender pay gap research is the motherhood penalty: women's wages decline after they have children, while men's wages often increase. This is not merely because mothers reduce hours or take career breaks -- it persists after controlling for those factors.

Correll's Audit Study

Sociologist Shelley Correll of Stanford University and colleagues conducted a landmark audit study published in the American Journal of Sociology (2007). They sent identical resumes to employers, varying only whether the fictitious applicant was identified as a parent. The results were striking:

Mothers were called back significantly less often than childless women with identical qualifications
Mothers were offered lower starting salaries when they were evaluated
Mothers were held to higher performance standards than non-mothers
Fathers were called back more often than childless men
Fathers were offered higher starting salaries than childless men

The evaluators were not applying a stated policy; they were making individual judgments about competence and commitment. The motherhood penalty and fatherhood bonus reflect implicit assumptions -- widely documented in experimental research by researchers including Amy Cuddy at Harvard and Susan Fiske at Princeton -- that mothers are less committed to work and that fathers are more stable and responsible employees.

Kleven's Child Penalty Research

Henrik Kleven of Princeton University and collaborators' research using comprehensive Danish administrative data -- tracking actual earnings for every worker in the country over time -- provided the most precise measurement of the child penalty. Published in the American Economic Review (2019), the study found that:

Women's earnings dropped by approximately 30% after the birth of a first child
Men's earnings showed virtually no change
The gap this created did not close over subsequent decades
Childless women earned nearly as much as childless men in Denmark
The child penalty accounted for the large majority of the remaining gender earnings gap

Kleven's finding -- that the gender pay gap is, to a large degree, a parenthood gap -- has been replicated in studies using administrative data from the United States (Kleven, Landais, and Sogaard, 2019), the United Kingdom (Costa Dias, Joyce, and Parodi, 2020), and several other countries. The magnitude of the child penalty varies across countries but is present everywhere.

"The gender pay gap is, to a large degree, a motherhood penalty." -- Henrik Kleven, Princeton University, 2019

What Discrimination Explains -- and What Remains Uncertain

The role of direct wage discrimination -- an employer paying a woman less than a man for genuinely identical work -- is genuinely difficult to measure, because it is impossible to observe "identical work" perfectly in non-experimental settings.

Evidence for Discrimination in Hiring

The evidence for discrimination in hiring and performance evaluation is substantial and well-replicated:

Resume audit studies consistently find gender gaps in callback rates in male-dominated fields. A 2012 study by Corinne Moss-Racusin and colleagues at Yale, published in Proceedings of the National Academy of Sciences, found that science faculty rated identical applications lower when the applicant had a female name, offered lower starting salaries, and were less willing to mentor the female-named applicant.

Orchestral auditions provide a natural experiment. A landmark study by Claudia Goldin and Cecilia Rouse, published in the American Economic Review (2000), found that switching to blind auditions (behind a screen) increased the probability that women advanced by 50% in preliminary rounds. This directly demonstrated that gender bias was operating in the evaluation process when evaluators could see the performers.

Performance evaluations in corporate settings show systematic gender differences. Research by Kieran Snyder (2014), analyzing 248 performance reviews, found that critical feedback directed at women was more likely to include personality-based criticism ("abrasive," "too aggressive") while critical feedback directed at men was more likely to focus on specific skills to develop.

The Honest Uncertainty

What remains contested is the share of the adjusted pay gap (the 2-8 cents after controlling for observable factors) that reflects ongoing wage discrimination versus unmeasured productivity-relevant differences. The honest answer is that both likely contribute, and the evidence does not allow a definitive decomposition. Economists call this the "unexplained residual" -- the gap that remains after all measurable factors are accounted for. It is an upper bound on discrimination (because some relevant factors may be unmeasured) but not a measure of zero discrimination (because discrimination is one plausible explanation for the residual).

International Comparisons: What Works

The gender pay gap varies substantially across countries, and this variation provides crucial evidence about what policies and institutions reduce it.

Country	Unadjusted Earnings Gap (OECD 2023)	Notable Policy Features
Belgium	~3.8%	Strong collective bargaining, pay transparency
Iceland	~10%	Pay transparency legislation (2018), mandatory equal pay certification
Denmark	~13%	Universal childcare, generous shared parental leave
United Kingdom	~14.5%	Gender pay gap reporting required for 250+ employee firms
United States	~17%	Limited federal policy; significant state variation
Canada	~16.7%	Federal pay equity legislation (2021)
Japan	~22%	Strong occupational segregation; limited women in management
South Korea	~31%	Highest in OECD; high occupational segregation, limited public childcare

Countries with smaller gender pay gaps consistently feature four institutional characteristics:

Affordable, accessible public childcare that allows mothers to maintain continuous labor force attachment. In Denmark, public childcare costs are capped at approximately 25% of total cost; in the United States, the average annual cost of center-based infant care exceeds $15,000, according to the Economic Policy Institute (2023).
Paid parental leave designed for both parents, particularly non-transferable paternity leave ("daddy quotas") that normalizes male caregiving. Iceland's parental leave system allocates 6 months to each parent plus 6 weeks that can be shared; uptake among fathers exceeds 90%.
Pay transparency requirements that expose and reduce within-firm pay gaps. Iceland's Equal Pay Standard (2018) requires companies with 25+ employees to obtain certification that they pay men and women equally for work of equal value.
Strong collective bargaining that compresses wage distributions and reduces the scope for individual negotiation -- a process in which gender differences in negotiation behavior (documented by researchers including Linda Babcock of Carnegie Mellon in her 2003 book Women Don't Ask) can compound into significant earnings gaps over time.

What Actually Reduces the Pay Gap: Evidence-Based Interventions

Pay Transparency

Mandatory pay transparency -- requiring employers to disclose pay information -- has been shown to reduce gender pay gaps in several natural experiments. Research by Morten Bennedsen and colleagues (2022), studying Denmark's 2006 pay transparency law, found that it reduced the gender pay gap by approximately 7% at affected firms.

The mechanism is information: women cannot negotiate toward market rates if they do not know what those rates are, and employers cannot suppress wages if doing so is publicly visible. The EU Pay Transparency Directive, adopted in 2023 and requiring implementation by 2026, will extend these requirements across the European Union.

Universal Childcare

The most powerful structural intervention in international data is affordable, high-quality public childcare. Quebec's 1997 introduction of universal low-cost childcare ($5/day, later $7/day) produced dramatic results: maternal labor force participation increased by 8 percentage points, and mothers' earnings increased substantially over the following decade, according to research by Pierre Lefebvre and Philip Merrigan (2008).

Paternity Leave That Men Actually Use

Several countries have experimented with non-transferable paternity leave that cannot be transferred to the mother. When fathers take extended parental leave, multiple things change: women's career interruption is shorter, domestic labor sharing becomes more equal, and fathers develop caregiving habits that reduce the premium employers can extract for temporal availability.

Research on Sweden's paternity leave reforms by Elly-Ann Johansson (2010) found that when fathers took more parental leave, mothers' long-term earnings were measurably higher -- approximately 6.7% higher income for each month of paternity leave taken. Norway's experience with "daddy quotas" showed similar effects.

Restructuring Greedy Jobs

Goldin's analysis implies that reducing the pay premium for extreme temporal inflexibility -- moving toward output-based compensation, reducing the penalty for predictable hours -- would mechanically reduce the within-occupation gender pay gap. The COVID-19 pandemic's forced experiment in remote work and flexible arrangements provides a natural experiment that researchers are beginning to analyze. Early evidence from the remote work culture shift suggests that flexible work arrangements may be narrowing within-firm gender pay gaps, though the data is still preliminary.

What the Data Shows -- and What It Does Not

The honest summary of gender pay gap research is:

The gap is real. Women earn significantly less than men in every country where data is available. The raw gap in the United States is approximately 16-18 cents; the adjusted gap is approximately 2-8 cents.
Multiple causes contribute. Occupational segregation, the motherhood penalty, discrimination in hiring and evaluation, the structure of "greedy jobs," negotiation differences, and historical devaluation of female-coded work all play documented roles.
The motherhood penalty is the largest single driver of the gap among prime-age workers, as demonstrated by Kleven's research using administrative data across multiple countries.
The adjusted gap is smaller but not zero. After controlling for measurable factors, a residual gap persists. Whether this reflects unmeasured discrimination, unmeasured productivity differences, or both is genuinely uncertain -- and honest analysis acknowledges this uncertainty rather than claiming certainty in either direction.
Structural solutions work. Countries that have invested in public childcare, shared parental leave, pay transparency, and strong collective bargaining show smaller gaps. This is not accidental -- it is evidence that the gap responds to institutional design.
The problem is not primarily about individual choices. The choices that appear to explain the gap -- which occupation to enter, how many hours to work, whether to take career breaks -- are themselves shaped by institutional structures, social norms, and labor market incentives that can be designed differently. Understanding this is crucial for moving beyond the sterile "choice vs. discrimination" debate that has dominated popular discussion.

References and Further Reading

Goldin, Claudia. (2021). Career and Family: Women's Century-Long Journey Toward Equity. Princeton University Press.
Goldin, Claudia. (2014). "A Grand Gender Convergence: Its Last Chapter." American Economic Review, 104(4), 1091-1119. doi.org/10.1257/aer.104.4.1091
Kleven, Henrik, Camille Landais, and Jakob Egholt Sogaard. (2019). "Children and Gender Inequality: Evidence from Denmark." American Economic Review, 109(7), 2748-2782. doi.org/10.1257/aer.20160440
Blau, Francine D. and Lawrence M. Kahn. (2017). "The Gender Wage Gap: Extent, Trends, and Explanations." Journal of Economic Literature, 55(3), 789-865.
Correll, Shelley J., Stephen Benard, and In Paik. (2007). "Getting a Job: Is There a Motherhood Penalty?" American Journal of Sociology, 112(5), 1297-1339.
Goldin, Claudia and Cecilia Rouse. (2000). "Orchestrating Impartiality: The Impact of 'Blind' Auditions on Female Musicians." American Economic Review, 90(4), 715-741.
Levanon, Asaf, Paula England, and Paul Allison. (2009). "Occupational Feminization and Pay: Assessing Causal Dynamics Using 1950-2000 U.S. Census Data." Social Forces, 88(2), 865-891.
Bertrand, Marianne, Claudia Goldin, and Lawrence F. Katz. (2010). "Dynamics of the Gender Gap for Young Professionals in the Financial and Corporate Sectors." American Economic Journal: Applied Economics, 2(3), 228-255.
Moss-Racusin, Corinne A., et al. (2012). "Science Faculty's Subtle Gender Biases Favor Male Students." Proceedings of the National Academy of Sciences, 109(41), 16474-16479.
Babcock, Linda and Sara Laschever. (2003). Women Don't Ask: Negotiation and the Gender Divide. Princeton University Press.
Bennedsen, Morten, et al. (2022). "Do Firms Respond to Gender Pay Gap Transparency?" Journal of Finance, 77(4), 2051-2091.
U.S. Census Bureau. (2024). Income and Poverty in the United States: 2023. census.gov
OECD. (2024). Gender Wage Gap (Indicator). data.oecd.org/earnwage/gender-wage-gap.htm
Economic Policy Institute. (2023). Child Care Costs in the United States. epi.org
Catalyst. (2024). Women in Management: Quick Take. catalyst.org

Frequently Asked Questions

What is the raw gender pay gap vs the adjusted gender pay gap?

The raw (unadjusted) gender pay gap compares the median earnings of all full-time male workers to all full-time female workers, regardless of occupation, industry, or hours worked. In the US, this is roughly 16-18 cents on the dollar. The adjusted gap controls for factors like occupation, industry, experience, and education, and typically falls to 2-8 cents depending on what is controlled for. Both measures are real; they answer different questions about where and how inequality operates.

What did Claudia Goldin find in her Nobel Prize-winning research?

Harvard economist Claudia Goldin, who won the 2023 Nobel Prize in Economics, found that the largest remaining driver of the gender pay gap is not discrimination within jobs but differences in how men and women are compensated for the same total hours worked. Jobs that pay premium wages for long, inflexible hours — law, finance, corporate management — disproportionately penalize career structures associated with caregiving. Goldin called this the 'greedy jobs' problem: the pay gap within occupations is largely driven by the nonlinear premium attached to availability and temporal flexibility.

What is the motherhood penalty?

The motherhood penalty refers to the wage and career penalties that women experience after having children, while men often experience a corresponding 'fatherhood bonus.' Research by Shelley Correll and others finds that mothers are rated as less competent and committed, offered lower salaries, and held to higher performance standards than identical childless women. Fathers are rated as more stable and committed and offered higher salaries than identical childless men. The effect accounts for a substantial portion of the gender pay gap among workers in their 30s and 40s.

What is occupational segregation and how does it affect the pay gap?

Occupational segregation refers to the clustering of men and women into different jobs and industries. Female-dominated occupations systematically pay less than male-dominated occupations requiring comparable education and skill. Research by Paula England and others finds that a significant portion of this pay difference cannot be explained by productivity-related factors — it reflects historical devaluation of work associated with women. As women enter previously male-dominated fields, those fields' relative pay often decreases, suggesting the pay is partly determined by who does the work.

What policies and practices most effectively reduce the gender pay gap?

Evidence-based approaches include: pay transparency (which reduces information asymmetries that allow individual negotiation gaps to compound); affordable, high-quality childcare (which allows women to maintain labor force attachment after children); flexible and shared parental leave (particularly paternity leave, which when used normalizes male caregiving and reduces the career cost concentrated on mothers); and restructuring jobs away from inflexible 'greedy' hour premiums toward output-based compensation. Countries with universal childcare and generous shared parental leave consistently show smaller gender pay gaps.