Four of the 40 companies in Germany’s DAX are run by a woman. Four of the 40 in France’s CAC 40. In the United States, 48 of the 500 in the S&P 500123. Three countries, three different approaches to the question, and much the same answer: about one company in ten.
The approaches really are different. France has required 40% of each sex on the boards of its listed companies since 2011, with no size threshold at all, and since 2021 has extended a 30% floor to their executive committees45. Germany has required 30% on the supervisory boards of 105 listed and co-determined companies since 20166. The United States has no federal quota, and the two rules that came closest are gone: California’s was struck down in 2022 and Nasdaq’s listing rule was vacated by a federal appeals court in 202478. One state rule survives, in Washington9.
On boards, the policy shows. France has 46.5% women on the boards of its largest listed companies and Germany 36.1%, the highest and the tenth highest of the 27 member states10. The obvious reading is that the quota worked, and it is partly right. The question is what else it did.
What the law did
France’s CAC 40 boards held 15.3% women in 2010. By 2014, when the law’s first threshold of 20% arrived, they held 30.0%. By 2017, when the 40% threshold arrived, 42.0%2. They have sat between 42% and 48% in the nine years since. Germany’s record is the same shape in miniature: the 105 companies bound by the fixed quota went from 25.0% in 2015 to 30.3% in 2016, the first year it applied, a gain of 5.3 points in 12 months. Companies not bound by it gained 1.4 points that year11.
Both numbers stopped where the statute stopped. The German government’s own evaluation of the law, carried out by consultants for the family ministry and laid before the Bundestag in 2020, found that the 30% minimum “derzeit in einem Großteil der Aufsichtsräte als Deckel wirkt” and that companies rarely tried to go beyond it12. The floor had become a ceiling.
The room the law can reach
Women's share of board seats at the largest listed companies, and of all managers, 2012 to 2025.
The data
| Year | France, boards | Germany, boards | France, all managers | Germany, all managers |
|---|---|---|---|---|
| 2012 | 21.4% | 14.9% | 39.3% | 28.8% |
| 2013 | 27.1% | 17.4% | 36.0% | 29.1% |
| 2014 | 29.6% | 19.6% | 32.7% | 29.0% |
| 2015 | 34.0% | 21.1% | 31.7% | 28.9% |
| 2016 | 40.0% | 24.1% | 32.9% | 29.5% |
| 2017 | 41.9% | 26.5% | 33.4% | 29.0% |
| 2018 | 42.7% | 28.1% | 34.4% | 29.2% |
| 2019 | 44.0% | 29.4% | 34.7% | 29.5% |
| 2020 | 43.9% | 29.8% | 35.5% | 28.6% |
| 2021 | 44.5% | 30.7% | 37.8% | 29.2% |
| 2022 | 44.8% | 32.8% | 39.9% | 28.9% |
| 2023 | 45.8% | 34.4% | 38.9% | 29.3% |
| 2024 | 47.1% | 35.9% | 39.5% | 29.5% |
| 2025 | 46.5% | 36.1% | 39.0% | 30.3% |
One caution belongs here. Germany’s quota-bound companies had already gone from 10.6% in 2009 to 22.6% in 2015, on a hand-collected count of the 111 firms the law would catch, before it bound anyone, while the measure was being debated in public13. Part of the 2016 jump was a trend that was already running. And France’s headline figure covers only the seats the quota governs, the directors elected at the annual meeting; employee representatives, who hold about 15% of CAC 40 board seats, were counted separately until this year2.
What it did not do
Outside the companies the law covers, almost nothing moved.
The share of women among all German managers was 28.8% in 2012 and 30.3% in 202514. The Federal Statistical Office, counting on its own definition, puts it more plainly: “Dieser Anteil veränderte sich seit 2012, dem Zeitpunkt der Einführung der aktuellen Klassifikation, nur wenig”15. On the narrower measure of senior and middle management, Germany gained a tenth of a point in 11 years16. France’s manager share was 36.4% in 1995 and 39.0% in 202517.
Inside those companies it did move. The supervisory-board quota took two years to reach the management board of the same firms, which went 8.4% in 2016, 8.5% in 2017 and 8.9% in 2018. After that it rose sharply: the 105 quota-bound companies went from 5.4% in 2015 to 17.9% in 2022, a gain of 12.5 points on the government’s own count, and the 65 firms now caught by the executive-board rule stood at 23.9% in late 2025, down from 24.3% a year earlier111. Most of that came before Germany’s executive-board rule bound a single appointment, which it did only from August 2022. The movement is real, and it stops at the edge of the companies the statute lists. France’s Haut Conseil à l’Égalité reported in 2019 that “le ruissellement attendu de ces instances de gouvernance vers les instances de direction ne s’est pas produit”, the expected trickle-down from governance bodies to management bodies having failed to arrive18.
The best-identified study of any of this is of Norway, which imposed a 40% board quota in 2003. Bertrand, Black, Jensen and Lleras-Muney found that the women appointed were better qualified than their female predecessors, and that the earnings gap inside the affected boards narrowed from about 36% to about 25%. Below the board they found nothing: no effect on the share of female employees, on women in the top five earners, or on women above the 90th earnings percentile in the firm. Their conclusion was that the reform had “very little discernible impact on women in business beyond its direct effect on the women who made it into boardrooms”19. Their most robust estimate was negative: raising the female board share from 5% to 40% cut the share of women in the firm’s top earnings quartile by about two points.
A working paper by Matsa and Miller, circulated in April 2026, reaches the same place from Germany’s 2021 law, which requires one woman on the management board of a listed and co-determined company with more than three board members20. It raised top-executive representation by about two thirds, drew the women in from outside existing networks and mostly into human resources and niche roles, produced no increase in female chief executives, and had “no significant effects on the female share of managers in lower ranks”21.
There is a reason to expect that law to reach no further. It binds 65 companies in the whole of Germany1. Of the 1,915 companies that owe any duty under the German statute, 683 of the 1,405 that published a target for their management board set that target at zero, and only a third of those gave the reasons the law requires11.
The rules came off
The United States is the control, and it did not behave as a control is supposed to.
Nasdaq’s rule, which asked listed companies to have at least one female director or explain why not, was in force from 6 August 2021 to 4 February 202522. California’s quota was unenforceable from May 2022. Over the rule’s own window, from the second quarter of 2021 to the second quarter of 2025, women’s share of Russell 3000 board seats rose from 25.6% to 30.1%, about 1.1 points a year. In the second quarter of 2026, the first full year after both rules were gone, it fell to 29.2%, which the organisation that tracks it calls the first annual decline in more than a decade23. Women’s share of new board appointments fell from 41.0% in 2022 to 22.7% in 2026. Among S&P 500 companies the share of new directors who were women fell from 43% in 2021 to 29% in 2026, and the proportion of boards disclosing a policy of considering diverse candidates collapsed from 58% to 12% in a single year3.
What happened when the rules went
Women's share of board seats in the Russell 3000, second quarter of each year. The band marks the period in which Nasdaq's board diversity rule was in force, 6 August 2021 to 4 February 2025.
The data
| Quarter 2 of | Women’s share of Russell 3000 board seats | Change |
|---|---|---|
| 2020 | 22.6% | – |
| 2021 | 25.6% | +3.0 pp |
| 2022 | 28.0% | +2.4 pp |
| 2023 | 29.0% | +1.0 pp |
| 2024 | 30.0% | +1.0 pp |
| 2025 | 30.1% | +0.1 pp |
| 2026 | 29.2% | −0.9 pp |
For comparison, women’s share of American management occupations has risen from 37.1% in 2003 to 41.8% in 2025: about 0.2 points a year, for 22 years, with no rule pointed at it24. The thing that moved quickly was the thing a rule was aimed at, and it moved back when the rule went.
One thing cuts against reading the American reversal that way. Germany’s numbers turned down over the same stretch with both its quotas in force, measured late in 2025 against late in 2024: the DIW’s annual count found the rise had largely stopped, the management boards of the 200 largest companies slipping from 19.1% to 18.6% and the DAX 40 from 25.9% to 25.4%1. A general retreat from diversity policy would show in both countries at once, and the American series on its own cannot tell that apart from the rules lapsing.
The court that ended Nasdaq’s rule did not hold that it was unconstitutional. Sitting as a full bench, the Fifth Circuit divided nine to eight and decided the case on statutory grounds: “We hold, however, that the diversity rules cannot be squared with the Securities Exchange Act of 1934.” A footnote records that the constitutional arguments were not reached8. California’s law fell on different ground, the equal protection clause of the state’s own constitution, and the state’s appeal has still not been decided7.
The measure everybody ranks by
America looks ahead of both quota countries on the headline figure, with 42.9% of managers against France’s 39.5% and Germany’s 29.5% in 2024, the last year all three are measured the same way14. That ranking does not survive examination.
The American category is far broader. Managers were 11.7% of all United States employment in 2025 and 4.4% of German employment, so the word is doing nearly three times as much work25. Within it, 79.6% of American human resources managers are women and 8.5% of construction managers24. Strip out the self-employed, who are about 28% of French managers and 9% of American ones, and France moves ahead of the United States, 44.4% against 44.0%. Among chief executives, senior officials and legislators, France is ahead again, 37.5% against 34.8%26.
The household survey that produces the American figure also reports that 33.0% of American chief executives are women, because it asks people what they call themselves. Among the Fortune 500, a different list counted on a different date, the figure is 11%27. Set the first number beside a European listed-company figure and the error is a factor of three.
Two cleaner comparisons run the other way. Prime-age female labour force participation is 77.8% in the United States, against 84.5% in Germany and 85.4% in France; the American rate was the highest of the three in 1990 and has gained under four points since28. And the American gender pay gap, on median full-time earnings, is the widest of the three at 17.3%, against Germany’s 13.5% and France’s 7.0%29.
Why Germany is last
Germany is last on every cut of the manager data, and has been flat on all of them for a decade. The reasons have little to do with company law.
Children explain 96.7% of Germany’s entire gender employment gap, and the German employment penalty for having a child, 41%, is 1.64 times the French figure of 25%30. On earnings the German penalty is 61% against 31% in the United States, on a specification that leaves France out31; French work on French data puts the penalty at around 30% in the five years after a birth, which is not the same measure32. Work by Kleven, Landais and Søgaard on Danish registers found the long-run earnings penalty “driven in roughly equal proportions by labor force participation, hours of work, and wage rates”, and it is the wage-rate and occupational-rank components that bear on who gets promoted33.
The German institutions that surround this are the most generous of the three and the least helpful. Germany gives mothers 37.6 weeks of leave at full-rate equivalent, France 20 and the United States none34. The country raised job protection for new mothers from two months to 36 months between 1979 and 1992. Schönberg and Ludsteck, studying those reforms, found that “the expansions had little impact on mothers’ labor force participation rates in the long run”, and that one month away from work lowers later wages by about 1%35. In 2001, 1.6% of German fathers took parental leave.
What Germany has instead
Three measures on which Germany is the outlier of the three countries. The child penalty is the long-run fall in a mother's employment relative to a father's.
The data
| Germany | France | United States | |
|---|---|---|---|
| Employment penalty for having a child | 41% | 25% | 25% |
| Women working part-time | 35.6% | 18.7% | 16.5% |
| Mothers' paid leave, full-rate-equivalent weeks | 37.6 weeks | 20.0 weeks | none |
Below that sit the hours. Formal childcare covered 60.9% of French under-threes in 2024 and 41.9% of German ones in 202336. German children who started school this August were the first with a legal entitlement to an eight-hour day, under a federal law passed for the purpose; it reaches the rest of primary school only as those cohorts move up37. A German mother of two is nearly eight times as likely to work part-time as a German father of two, 70.8% against 9.0%38. Across the whole female workforce the part-time rate is 35.6% in Germany, 18.7% in France and 16.5% in the United States39.
The sharpest evidence that company law is not the binding constraint sits inside Germany. The IAB’s establishment panel measures leadership representation as the share of women in management divided by their share of employees. In western Germany the second management level scores 0.93. In eastern Germany it scores 1.00, which is to say there is no gap at all, under the same company law and the same quota. The IAB does not say why, and the two halves of the country differ in their industries and in the size of their firms as well as in their childcare40.
What nobody knows
Three things in this story are unsettled.
1. Nobody has decomposed women’s share of leadership across a whole economy the way the wage gap has been decomposed, so no share of the gap in any of these three countries can be assigned to a cause. Within a single firm it has been done, and the promotion study below is one. Blau and Kahn’s standard survey of the wage gap leaves 38% unexplained once occupation, industry and union status are controlled, and warns that the residual may be compensating differentials as readily as discrimination, while the controls may themselves be its products41. That residual is a ceiling on discrimination, not a measurement of it. Their finding most relevant here is that the gap is largest at the top and has fallen slowest there.
2. Whether women at the top pull others up is contested by method. Kunze and Miller, using Norwegian employer-employee records, find that a one-standard-deviation rise in the share of female bosses would narrow the promotion gap by 13.8%, while the same rise in the share of female peers would widen it by 52.7%42. The quota studies, which have the better identification, find nothing. No reconciliation of the two has been published, as far as this paper could find.
3. The evidence on discrimination points in different directions depending on where it looks. A meta-analysis of 57 hiring audit studies by Galos and Coppock finds no average bias against women at all, and instead a gradient by occupation: about three points against women in an all-male occupation and five points in their favour in an all-female one43. Those studies measure first-stage callbacks, and few of them reach senior posts. Where the evidence is strongest is promotion. Benson, Li and Shue, following 29,809 management-track employees of an American retailer, found women 7.4% more likely than men to receive the top rating for current performance and 12% more likely to receive the lowest rating for future potential, which accounts for between a third and 55% of a promotion gap of 1.64 percentage points44. The complication they report themselves is that potential ratings do carry real information about later performance, so discarding them would close the gap and make promotions worse.
What is not in doubt is the shape of the thing. A quota is an instrument of unusual precision: it fills the seats it names, in the companies it names, and the evidence that it does so is as good as evidence in economics gets. It is also an instrument of almost no reach. Germany has spent 10 years legislating about 105 boardrooms, and the share of women among its 1.9 million managers has moved by less than two points.
Sources
-
Arianna Antezza, Alina Meiner and Katharina Wrohlich, “Frauen in Vorständen und Aufsichtsräten großer Unternehmen: Aufwärtstrend der letzten Jahre kommt größtenteils zum Halt”, DIW Wochenbericht 3/2026, 14 January 2026, pp. 17–29, doi:10.18723/diw_wb:2026-3-2. The DIW Managerinnen-Barometer; data collected 3 November to 5 December 2025. Four of the 40 DAX companies had a woman as chief executive. ↩︎ ↩︎ ↩︎ ↩︎
-
Institut Français des Administrateurs and Ethics & Boards, “IFA – Ethics & Boards Barometer of Gender Diversity in Leadership Bodies”, CAC 40 / SBF 120, 4th edition, February 2026, pp. 10, 21–23, 27 and the glossary at p. 32. A private data provider’s product, not an official statistic; it counts voting directors elected at the annual general meeting. ↩︎ ↩︎ ↩︎
-
Spencer Stuart, “2026 US Spencer Stuart Board Index”, September 2026, pp. 17, 22–24 and 47. Based on the proxy statements of 490 S&P 500 companies filed to 30 April 2026. ↩︎ ↩︎
-
Loi n° 2011-103 du 27 janvier 2011 relative à la représentation équilibrée des femmes et des hommes au sein des conseils d’administration et de surveillance et à l’égalité professionnelle, consolidated in the Code de commerce at articles L22-10-3, L22-10-21, L225-18-1, L225-69-1 and L226-4-1. The sanctions are the nullity of a breaching appointment and, under article L225-45, the suspension of directors’ remuneration until the board is lawful. Legifrance refuses automated requests; the consolidated articles were read from the Legifrance mirror at codes.droit.org. ↩︎
-
Loi n° 2021-1774 du 24 décembre 2021 visant à accélérer l’égalité économique et professionnelle, article 14, which inserts article L23-12-1 into the Code de commerce. Companies of 1,000 employees or more must reach 30% of each sex among their senior executives and the members of their leadership bodies from 1 March 2026, and 40% from 1 March 2029; the financial penalty of up to 1% of payroll applies only from the later date. ↩︎
-
Aktiengesetz § 96, paragraph 2, in force since 1 May 2015 and applicable to supervisory-board elections from 1 January 2016: “Bei börsennotierten Gesellschaften, für die das Mitbestimmungsgesetz, das Montan-Mitbestimmungsgesetz oder das Mitbestimmungsergänzungsgesetz gilt, setzt sich der Aufsichtsrat zu mindestens 30 Prozent aus Frauen und zu mindestens 30 Prozent aus Männern zusammen.” An election in breach is void, so the seat stays empty. ↩︎
-
California Corporations Code § 301.3, added by Senate Bill 826 of 2018, required one woman on the board by the end of 2019 and up to three by the end of 2021, on penalty of $100,000 for a first violation and $300,000 thereafter; the text is reproduced at california.public.law (the legislature’s own site refuses automated requests). The section was held to violate the equal protection clause of the California Constitution in Crest v. Padilla, Los Angeles County Superior Court No. 19STCV27561, judgment of 13 May 2022, and has been unenforceable since. This paper could not obtain the judgment itself or an appellate case number; that the state’s appeal was still undecided in July 2026 is reported by Judicial Watch, which acted for the plaintiffs and is not a neutral source. ↩︎ ↩︎
-
Alliance for Fair Board Recruitment v. Securities and Exchange Commission, No. 21-60626, United States Court of Appeals for the Fifth Circuit, en banc, 11 December 2024, at 2 and footnote 2. The court divided nine to eight, Judge Oldham writing for the majority. Footnote 2 records that the constitutional challenges were not reached. ↩︎ ↩︎
-
Revised Code of Washington 23B.08.120, enacted as 2020 c 194 s 1 and in force for annual meetings from 1 January 2022, read 10 November 2026. A Washington public company must either have a “gender-diverse board”, meaning that women were “at least twenty-five percent of the directors” for at least 270 days of the preceding financial year, or deliver a board diversity discussion to its shareholders instead. The only remedy is a shareholder application to compel that disclosure, so it is the same comply-or-explain form as Nasdaq’s rule rather than a quota. ↩︎
-
Eurostat, sdg_05_61, “Positions held by women in senior management”, which republishes the European Institute for Gender Equality’s Women and Men in Decision-Making database. Board members of the largest listed companies, at most 50 per country, second half-year. Eurostat last update 25 March 2026, retrieved 10 November 2026. ↩︎
-
Bundesregierung, ninth annual information on the share of women in management levels and bodies of the private sector and the federal public service, Bundestags-Drucksache 21/2300, 16 October 2025, figures 1, 2 and 32 and the tables on the Zielgrößen. The figures run to financial year 2022 for all 1,915 reporting companies and to 2023 for listed ones. The same document reports that only 55 of the 1,915 companies disclosed the actual composition of the first management level, so it carries no usable figure for the levels below the board. ↩︎ ↩︎ ↩︎
-
Bundesregierung, evaluation report on the effectiveness of the Act on equal participation of women and men in leadership positions, together with the government’s response, Bundestags-Drucksache 19/24615, 19 November 2020, pp. 7–9 and 24. The evaluation was carried out by Kienbaum Consultants International. It also found that works councils, human resources departments, managers and employees alike “haben häufig kaum oder keine Kenntnis von der strategischen Vorgehensweise zur Festlegung und den tatsächlich festgelegten Zielgrößen des eigenen Unternehmens”. ↩︎
-
Viktor Bozhinov, Christopher Koch and Thorsten Schank, “The Second Glass Ceiling: Women’s Role in Supervisory Boards of German Firms”, Schmalenbach Business Review 71(3), pp. 385–411, doi:10.1007/s41464-018-0063-1; read in the IZA discussion paper 11057 version of September 2017. Hand-collected data on every supervisory director of the 111 firms bound by the German quota, 2009 to 2016. The same study finds no concentration of appointments in a few women: mean simultaneous mandates among shareholder representatives were 1.48 for men and 1.47 for women in 2009, and 1.38 for men against 1.21 for women in 2016. ↩︎
-
International Labour Organization, ILOSTAT, dataflow ILO,DF_SDG_T552_NOC_RT,1.0, United Nations Sustainable Development Goal indicator 5.5.2, the proportion of women in managerial positions, defined on all of major group 1 of the occupational classification and covering the self-employed as well as employees. Retrieved 10 November 2026. The 2024 values are Germany 29.48%, France 39.52% and the United States 42.86%. The American figures are for ages 16 and over and exclude the armed forces, the European ones for 15 and over. ↩︎ ↩︎
-
Statistisches Bundesamt, “Frauen in Führungspositionen”, indicator page dated 23 July 2026, retrieved 10 November 2026. Women as a share of all employed persons in major group 1 of the International Standard Classification of Occupations, from the labour force survey. Destatis notes breaks in the series in 2005, 2011, 2012, 2020 and 2021. ↩︎
-
International Labour Organization, ILOSTAT, dataflow ILO,DF_SDG_0552_NOC_RT,1.0, senior and middle management, which is major group 1 of the occupational classification less sub-major group 14. Germany 27.58% in 2013 and 27.69% in 2024. Retrieved 10 November 2026; ILO last update 5 November 2026. ↩︎
-
Eurostat, lfsa_egais, employment by sex, age and occupation, own calculation of women as a share of all employed persons in major group 1. Last update 30 October 2026, retrieved 10 November 2026. The French series carries breaks at 2011, when the occupational classification changed, in 2013 and 2014, and in 2021; the fall to 31.8% in 2015 is a coding discontinuity rather than an event. ↩︎
-
Haut Conseil à l’Égalité entre les femmes et les hommes, Accès des femmes aux responsabilités et rôle levier des financements publics, rapport n° 2019-12-12-PAR-41, 12 December 2019, p. 19: “le ruissellement attendu de ces instances de gouvernance vers les instances de direction ne s’est pas produit”. The same page gives the share of women on the boards of the largest listed French companies as 10.9% in 2009 and 44.6% in 2019, against 7.3% and 18.2% on their executive committees. ↩︎
-
Marianne Bertrand, Sandra E. Black, Sissel Jensen and Adriana Lleras-Muney, “Breaking the Glass Ceiling? The Effect of Board Quotas on Female Labour Market Outcomes in Norway”, The Review of Economic Studies 86(1), 2019, pp. 191–239, doi:10.1093/restud/rdy032. Quotations and estimates read in the NBER working paper 20256 version, pp. 4, 22–26 and tables 5 to 7: “Overall, seven years after the board quota policy fully came into effect, we conclude that it had very little discernible impact on women in business beyond its direct effect on the women who made it into boardrooms.” ↩︎
-
Aktiengesetz § 76, paragraph 3a, in force since 12 August 2021 and applicable to appointments from 1 August 2022: where the management board of a listed and parity-co-determined company “aus mehr als drei Personen, so muss mindestens eine Frau und mindestens ein Mann Mitglied des Vorstands sein. Eine Bestellung eines Vorstandsmitglieds unter Verstoß gegen dieses Beteiligungsgebot ist nichtig.” The rule is a headcount, not a percentage. ↩︎
-
David A. Matsa and Amalia R. Miller, “A Gender Quota for Top Executives: Diversity without Disruption”, NBER working paper 35030, April 2026, abstract. A working paper, not yet refereed: “The quota increased female representation among top executives by about two-thirds”, and “We find no significant effects on the female share of managers in lower ranks, policies promoting gender equality, firm value, or performance.” ↩︎
-
Securities and Exchange Commission, Release No. 34-102281, File No. SR-NASDAQ-2025-007, 24 January 2025, 90 Fed. Reg. 8554, repealing Nasdaq’s board diversity listing requirement with effect from 4 February 2025, the effective date of the court’s decision. The rules had been approved on 6 August 2021 by Release No. 34-92590. ↩︎
-
50/50 Women on Boards, “2026 State of Board Leadership”, Gender Diversity Index Report Q2 2026, 6 October 2026, pp. 3–4 and 8, on Equilar data for the 2,979 companies of the Russell 3000. An advocacy organisation’s report: it describes the fall to 29.2% as the first annual decline in more than a decade, and adds that the gains of the previous five years “are not self-sustaining”. ↩︎
-
Bureau of Labor Statistics, Current Population Survey, table 11, employed persons by detailed occupation and sex, annual averages, retrieved 10 November 2026, with earlier years from the same table at
bls.gov/cps/aa<year>/aat11.txt. The narrow line is “Management occupations”; the broader “Management, business, and financial operations occupations” also contains accountants, analysts and human resources specialists. The 2025 averages cover 11 months: October 2025 was not collected because of the federal government shutdown. The bureau records breaks in the occupational classification in January 2003 and January 2011 and states that data from 2011 “are not strictly comparable with earlier years”. ↩︎ ↩︎ -
Own calculation from International Labour Organization, ILOSTAT, dataflow ILO,DF_EMP_TEMP_SEX_OCU_NB,1.0, employment by sex and occupation, retrieved 10 November 2026: managers as a share of total employment in 2025 were 11.70% in the United States, 7.42% in France and 4.42% in Germany, and Germany had 1.879 million of them. The self-employed were 28% of French managers and 9% of American ones in 2024, from the same flow set against the employees-only counts below. ↩︎
-
Own calculation from International Labour Organization, ILOSTAT, dataflow ILO,DF_EES_TEES_SEX_OC2_NB,1.0, employees by sex and two-digit occupation, retrieved 10 November 2026. Counting employees only, women were 44.44% of managers in France and 44.01% in the United States in 2024; in sub-major group 11, chief executives, senior officials and legislators, 37.5% in France and 34.8% in the United States. ↩︎
-
Fortune, “Women run a record 11% of Fortune 500 companies in 2026”, 3 June 2026: 55 of the 500. The count of 55 was first reached on the 2025 list, so the 2026 figure equals rather than exceeds it. The historical series is Catalyst, “Historical List of Women CEOs of the Fortune Lists: 1972–2023”, June 2023, which notes that the list covered only industrial companies until 1994. ↩︎
-
International Labour Organization, ILOSTAT, dataflow ILO,DF_EAP_DWAP_SEX_AGE_RT,1.0, labour force participation rate by sex and age, women aged 25 to 54, from national labour force surveys rather than modelled estimates. Retrieved 10 November 2026. In 2025 the rate was 84.47% in Germany, 85.39% in France and 77.80% in the United States; in 1990 it was 65.59%, 72.20% and 74.00%. The German step between 1990 and 1991 reflects reunification. ↩︎
-
OECD, gender wage gap, the difference between median earnings of men and of women relative to median earnings of men, full-time employees, retrieved 10 November 2026. In 2024 the gap was 17.29% in the United States, 13.47% in Germany and 7.00% in France. The French series is four-yearly to 2022 rather than annual. ↩︎
-
Henrik Kleven, Camille Landais and Gabriel Leite-Mariante, “The Child Penalty Atlas”, Review of Economic Studies 92(5), 2024, pp. 3174–3207, doi:10.1093/restud/rdae104; figures from the project’s own data export, retrieved 10 November 2026. The long-run employment penalty is 41% in Germany, 25% in France and 25% in the United States, and children account for 96.7% of Germany’s gender employment gap, 88.7% of France’s and 84.5% of America’s. The German and French estimates rest on Luxembourg Income Study microdata ending in 2018. ↩︎
-
Henrik Kleven, Camille Landais, Johanna Posch, Andreas Steinhauer and Josef Zweimüller, “Child Penalties across Countries: Evidence and Explanations”, AEA Papers and Proceedings 109, 2019, pp. 122–126, doi:10.1257/pandp.20191078; read in the NBER working paper 25524 version. The long-run earnings penalties on one specification are 61% in Germany, 51% in Austria, 44% in the United Kingdom, 31% in the United States, 27% in Sweden and 21% in Denmark. France is not among the countries estimated on this specification. ↩︎
-
Dominique Meurs and Pierre Pora, “Gender Equality on the Labour Market in France: A Slow Convergence Hampered by Motherhood”, Economie et Statistique / Economics and Statistics 510–511–512, 2019, pp. 109–130, doi:10.24187/ecostat.2019.510t.1990. Their figure is quoted here at one remove, from Stephen Bazen, Xavier Joutard and Hélène Périvier, IZA discussion paper 14763, October 2021, pp. 2 and 3, which reports that “studies suggest a child penalty for mothers of around 30% over the five years following a birth” on administrative data for 2005 to 2015; the Insee article itself was not reachable from here. Measured over five years after a birth, it is not on the same basis as the long-run penalties above. ↩︎
-
Henrik Kleven, Camille Landais and Jakob Egholt Søgaard, “Children and Gender Inequality: Evidence from Denmark”, American Economic Journal: Applied Economics 11(4), 2019, pp. 181–209, doi:10.1257/app.20180010; quotation from the NBER working paper 24219 version, p. 1. The same study puts the long-run penalty on occupational rank at 6% and on the probability of being a manager at 1.32 percentage points, and finds that children came to account for about 80% of all Danish gender earnings inequality by 2013, against about 40% in 1980. ↩︎
-
OECD Family Database, indicator PF2.1, parental leave systems, tables PF2.1.A and PF2.1.B, the entitlement in force in 2026, downloaded 10 November 2026. Expressed in full-rate-equivalent weeks, which discount a longer leave paid at a lower rate, mothers’ entitlement is 37.6 weeks in Germany, 20.0 in France and none in the United States; fathers’ is 4.7, 8.1 and none. ↩︎
-
Uta Schönberg and Johannes Ludsteck, “Expansions in Maternity Leave Coverage and Mothers’ Labor Market Outcomes after Childbirth”, Journal of Labor Economics 32(3), 2014, pp. 469–505, doi:10.1086/675078; quotations and estimates read in the IZA discussion paper 2699 version of March 2007, pp. 3, 5 and 27, so the published figures may differ. German job-protected leave was lengthened from two months to six in 1979, to 10 in 1986, 15 in 1989, 18 in 1990 and 36 in January 1992. The figure of 1.6% of fathers taking parental leave in 2001 is given there on the authority of Engstler and Menning (2003). ↩︎
-
OECD Family Database, indicator PF3.2, enrolment in childcare and pre-school, sheet for children aged 0 to 2, downloaded 10 November 2026. France 60.9% in 2024 and Germany 41.9% in 2023; the newest American figure in the same table is 27.97%, for 2011, which is why the United States is left out of the comparison. ↩︎
-
Sozialgesetzbuch VIII § 24. Paragraph 2 has given a child an entitlement to a day-care place from its first birthday since 1 August 2013. Paragraph 4 gives a child entering the first class of school in the 2026/2027 school year or later an entitlement to day-care provision of eight hours on working days, until the fifth class. ↩︎
-
Eurostat, lfst_hhptechi, part-time employment of adults by sex, age and number of children, women and men aged 18 to 64, Germany, 2025, retrieved 10 November 2026. Among women with two children 70.8% worked part-time, against 9.0% of men with two children; among women whose youngest child was under six, 67.6%. ↩︎
-
OECD, incidence of part-time employment, part-time employees as a share of all employees of that sex on the harmonised definition of under 30 usual weekly hours in the main job, retrieved 10 November 2026. Among women in 2025 the rate was 35.58% in Germany, 18.72% in France and 16.52% in the United States; in 1990 it was 29.59%, 22.83% and 20.22%. ↩︎
-
Susanne Kohaut, Iris Möller and Michael Oberfichtner, “Gender-Leadership-Gap in Deutschland”, IAB-Kurzbericht 24/2025, 4 November 2025, doi:10.48720/IAB.KB.2524, and the establishment-panel table for 2004 to 2022. Private-sector establishments only. Women held 29% of first-level and 42% of second-level management positions in 2024, against 25% and 33% in 2004, while their share of all employees rose from 41% to 45%. ↩︎
-
Francine D. Blau and Lawrence M. Kahn, “The Gender Wage Gap: Extent, Trends, and Explanations”, Journal of Economic Literature 55(3), 2017, pp. 789–865, doi:10.1257/jel.20160995; estimates read in the NBER working paper 21913 version. On Panel Study of Income Dynamics data for 1980 to 2010, 38% of the gap is unexplained once education, experience, region, race, occupation, industry and union status are controlled, and 85% with human capital alone; the share attributable to human capital fell from 27% to 8% while occupation and industry rose from 20% to 51%. ↩︎
-
Astrid Kunze and Amalia R. Miller, “Women Helping Women? Evidence from Private Sector Data on Workplace Hierarchies”, The Review of Economics and Statistics 99(5), 2017, pp. 769–775, doi:10.1162/rest_a_00668; estimates read in the NBER working paper 20761 version. Norwegian employer-employee records across seven ranks. The design is observational, and the quota studies that find nothing are the better identified. ↩︎
-
Diana Roxana Galos and Alexander Coppock, “Gender composition predicts gender bias: A meta-reanalysis of hiring discrimination audit experiments”, Science Advances 9(18), 2023, eade7979, doi:10.1126/sciadv.ade7979. 57 studies and 298 study-occupation estimates: the average treatment effect is 1.2 percentage points in women’s favour, and the gradient by the share of women already in the occupation is 8.15, implying 3.02 points against women in an all-male occupation and 5.13 in their favour in an all-female one. A second meta-analysis of 37 American studies reaches an insignificant pooled estimate: So Yun Park and Eunsil Oh, “Getting a Foot in the Door: A Meta-Analysis of U.S. Audit Studies of Gender Bias in Hiring”, Sociological Science 12, 2025, pp. 26–50. ↩︎
-
Alan Benson, Danielle Li and Kelly Shue, “‘Potential’ and the Gender Promotion Gap”, American Economic Review 116(2), 2026, pp. 375–417, doi:10.1257/aer.20220831; figures read in the working-paper version of 22 June 2022, pp. 2, 15 and tables 3 and 11, so the published figures may differ. The annual promotion rate is 10.98% for women against 12.62% for men. A one-point higher potential rating predicts a performance rating about a quarter of a standard deviation higher a year later, which is why discarding the ratings would both close the gap and lower the quality of promotions. ↩︎
.jpg){kind=link}