Claim: “If male variability theory is true, we would expect mostly men in highest positions in all industries and government”

Accuracy Assessment: Largely True

Important framing note: This claim is explicitly conditional — it assumes the Greater Male Variability Hypothesis (GMVH) is true and asks whether the stated consequences logically follow. The assessment therefore evaluates the logical validity of the inference, not the truth of GMVH itself.

Supporting article: The foundational claim that GMVH is empirically true is assessed separately: Men vary more than women in range of intellectual and physical abilities.

The core logical inference — that greater male variance in cognitive ability, risk preferences, and drive would produce more men at extreme performance tails, and therefore more men in the highest positions under merit-only selection — is mathematically and logically sound. Greater variance in a distribution necessarily means more members of that group appear at both extremes. This is not controversial. If GMVH is true, the prediction that men are overrepresented at the highest performance levels does follow as a direct mathematical consequence.

The empirical data is broadly consistent with this prediction. Men hold ~89.6% of Fortune 500 CEO positions. Only 13 of 193 UN member countries (6.7%) are currently led by women, and only 63 of 193 have ever had a woman head of government. In objectively evaluated domains (chess grandmasters: ~97.8% male; Nobel Physics: ~97% male; Fields Medal in mathematics: ~97% male — 2 of ~64 recipients), men dominate overwhelmingly — and in some domains at levels exceeding what cognitive GMVH alone would arithmetically predict.

Furthermore, multi-trait GMVH predicts that this dominance should result in only a low percentage of women reaching the apex of competitive hierarchies. Applying variance ratios of 1.14–1.25 across 3–4 relevant leadership traits simultaneously, the predicted female fraction at extreme combined thresholds falls to approximately 5–14% — entirely consistent with the ~10% female Fortune 500 CEOs and ~7% female heads of government observed empirically. The only overstatement in the original claim is “all industries”: in fields where GMVH-relevant traits are not the primary success factors (e.g., healthcare, education, social work), women already lead without any predicted GMVH disadvantage. The overall assessment is Largely True.

Key Claims at a Glance

Claim Breakdown

1. “GMVH logically predicts more men at the top of any trait distribution”

✅ True — the mathematical logic is sound

If GMVH is assumed true (as the claim instructs), the logical consequence is direct and mathematically unavoidable: when one group has greater variance in a trait, more members of that group will appear at both extremes of the distribution, including the top.

This is not speculation — it is a basic property of normal distributions. Consider two distributions with the same mean but different standard deviations. The distribution with the larger standard deviation (men, under GMVH) will necessarily contain a higher proportion of its members at any given extreme threshold. The further out the threshold, the larger the male-to-female ratio at that tail.

Mathematical illustration (with variance ratio = 1.14, i.e., males 14% more variable as per Baye & Monseur 2016):

The Baye & Monseur (2016) study, examining 12 large-scale international assessment databases (PISA, PIRLS, TIMSS, 1995–2015) across dozens of countries, explicitly concluded: “The ‘greater male variability hypothesis’ is confirmed.” Males were overrepresented at both extremes across mathematics, reading, and science.

The Heterodox Academy notes: “At the top end of any distribution of test scores where men have higher variability, we’d expect men to make up more than 50% of the upper end of the tail. Thus, any company drawing from the top 5% is likely to find a pool that contains more males. As one goes further out into the tail (i.e. becomes even more selective) the gender tilt becomes larger.”

Verdict: ✅ True. The logical inference is mathematically sound — if GMVH is true, more men appear at the extreme upper tails of any relevant trait distribution.

2. “GMVH applies to relevant traits beyond IQ — including risk appetite, drive, competitiveness”

✅ True — GMVH extends to economic and social preferences

Leadership positions at the highest levels require more than cognitive ability. They require:

• Willingness to take extreme career risks
• Extreme time preference (willingness to sacrifice present comfort for future reward)
• Competitive drive and status-seeking
• Tolerance for high-stakes decision-making under uncertainty

The Volk, Thöni & Ruigrok (2021) PNAS meta-analysis (50,000+ individuals, 97 samples) found converging evidence for greater male variability in all three of these preference categories:

• Risk preferences: men more variable (variance ratio ~1.25)
• Time preferences (patience/impatience): men more variable
• Social preferences (altruism, trust, competitiveness): men more variable

The significance statement in the paper: “The current literature underestimates the importance of gender differences and their effects on differential choices and outcomes between women and men” — because it focuses on mean differences and ignores the compounding effects of variance differences.

This is the critical insight: even if the mean difference between men and women in each trait is small, combined greater variance across multiple relevant traits compounds significantly. A person in the top 0.01% for cognitive ability AND risk tolerance AND competitive drive is far more likely to be male if GMVH applies to all three traits independently.

Verdict: ✅ True. GMVH extends well beyond IQ to the economic preferences most directly relevant to pursuing and succeeding in top leadership positions.

3. “Empirical data shows men dominating the highest positions across most industries and government”

✅ True — men dominate the highest positions in most sectors

The empirical picture is consistent: men vastly outnumber women at the top of most industries and in government leadership worldwide.

Corporate leadership (USA): | Metric | Women | Men | |——–|——-|—–| | Fortune 500 CEOs (2024) | 52 (10.4%) | 448 (89.6%) | | S&P 500 CEOs (2023) | 41 (8.2%) | 459 (91.8%) |

Fortune reports that women crossing the 10% threshold of Fortune 500 CEOs in 2023 was described as a historic milestone — after 70 years of the list’s existence.

Government (worldwide): | Metric | Women | Men | |——–|——-|—–| | Current heads of government (2024) | ~13 (6.7%) | ~180 (93.3%) | | Countries that have EVER had a woman leader | 63 (32.6%) | 130 (67.4%) | | Countries that have NEVER had a woman leader | 113 (58.5%) | — |

Source: Pew Research Center / Council on Foreign Relations Women’s Power Index

Objectively evaluated cognitive domains (no selection bias possible): | Domain | Women | Men | |——–|——-|—–| | Chess Grandmasters (FIDE) | 37 (2.2%) | 1,643 (97.8%) | | Nobel Physics 1901–2020 | ~3% | ~97% | | Fields Medal (mathematics) | 2 of ~64 (~3%) | ~97% |

Chess is a particularly robust natural experiment because ratings are entirely objective — they reflect competitive wins and losses with no subjective human evaluation. Discrimination cannot inflate or suppress ratings. The near-complete male dominance of chess grandmasters represents meritocratic outcome data in a cognitively demanding domain.

Verdict: ✅ True. Men dominate the highest positions across most industries and in government worldwide, at levels broadly consistent with — and in some domains exceeding — what GMVH would predict.

4. “‘In all industries’ is accurate — men dominate the top in every sector”

❌ Mostly False — men do not dominate all industries at the top

The claim states “all industries and government.” This is an overstatement.

There are industries and domains where women are well-represented or even dominant at the professional level, including:

• Healthcare: Women constitute the majority of nurses, midwives, and health workers globally; in many countries they represent 50%+ of physicians
• Education: Women dominate primary and secondary school teaching in most developed countries and hold substantial representation in educational administration
• Social work, psychology, counselling: Women constitute the majority of practitioners
• Publishing, journalism, public relations: Women hold substantial or majority representation
• Human resources: Women are the majority at all levels including senior HR roles in most countries

While men still dominate top leadership in several of these fields (hospital CEOs, university presidents), the gap is far smaller and in some cases absent compared to finance, tech, defence, and politics.

Furthermore, the GMVH-based prediction is strongest for traits most relevant to competitive, status-hierarchical domains. In fields where success depends more on conscientiousness, agreeableness, and communication — traits on which women score comparably or higher on average — GMVH-based male advantage is diminished.

Verdict: ❌ Mostly False. The “all industries” formulation is too broad. Men dominate the highest positions in most industries, but not all. Several sectors show near-parity or female dominance at various levels without DEI intervention.

5. “GMVH predicts only a low percentage of women at the top of business, industry and management”

✅ True — multi-trait GMVH predicts low but non-zero female representation at extreme thresholds

When GMVH variance ratios are applied across the multiple traits most relevant to reaching the apex of competitive hierarchies — cognitive ability, risk tolerance, competitive drive, and status-seeking — the combined compounding effect strongly constrains the predicted female fraction. Top leadership positions are not selected on one trait in isolation; they require extreme performance across several simultaneously.

Multi-trait GMVH prediction table (assuming independent traits, equal means):

Methodology: For each trait, a threshold t is set at the top 0.1% of the female distribution (t ≈ 3.09 standard deviations). Because male variance is higher (σ_m = √VR × σ_f), the same threshold corresponds to a proportionally smaller male z-score, so more males clear it. The fraction of successful candidates who are male is P_male^N / (P_male^N + P_female^N), where N is the number of traits. Variance ratios are drawn from Baye & Monseur (2016) (VR = 1.14 for academic performance) and Volk et al. (2021) (VR ≈ 1.25 for economic preferences); equal means assumed throughout.

Under a 3–4 trait model with VR in the range of 1.14–1.25, GMVH predicts approximately 5–14% female representation at the extreme combined performance threshold relevant to top leadership selection. This is a low percentage relative to population proportions (~50% female), and it aligns closely with empirical observation:

The prediction of “only a low percentage” is confirmed as mathematically precise: under GMVH, this low percentage is not zero — some women will always clear high multi-trait thresholds — but it is well below population proportions and constitutes a small minority figure entirely consistent with what the data shows.

Important caveats on the multi-trait model:

This calculation makes simplifying assumptions. In reality:

• The relevant traits are correlated, not independent — for example, cognitive ability and risk tolerance typically correlate at r ≈ 0.1–0.3 across the general population (see e.g. Dohmen et al., 2010, American Economic Review). Positive between-trait correlations reduce the compounding effect, meaning the true multi-trait prediction falls somewhere between the single-trait value (~35% female) and the fully-independent multi-trait floor (~2–8% female). A rough sensitivity check suggests moderate within-person trait correlations (r ≈ 0.3) shift the 3-trait prediction from ~14% toward ~20–25% female.
• The threshold for reaching top leadership is not purely trait-based; access to networks, mentorship, and opportunity also matter.
• GMVH variance ratios are measured at the population level and may not apply identically at all points of the distribution.

Even taking these caveats into account, the corrected prediction range (~5–25% female depending on model assumptions) still constitutes a “low percentage” relative to population proportions, and the empirical figures fall well within it.

Verdict: ✅ True. Multi-trait GMVH predicts approximately 5–14% female representation at extreme combined performance thresholds (widening to ~5–25% under correlated-trait assumptions). This is unambiguously a low percentage relative to the 50% population base. The observed figures — ~10% Fortune 500 CEOs, ~7% heads of government, ~2–3% in objectively scored cognitive domains — fall squarely within this predicted range.

6. “The level of male dominance observed is consistent with GMVH predictions”

🟡 Partially — GMVH explains the direction but not always the full magnitude

The empirical levels of male dominance observed (90%+ CEOs, 93%+ heads of state, 97%+ chess grandmasters) require careful examination against what GMVH’s variance ratios would mathematically predict.

What GMVH variance ratios predict at extreme tails (single trait):

With a variance ratio of 1.14 (Baye & Monseur 2016, for academic performance):

With the higher risk-preference variance ratio of 1.25 (Volk et al. 2021):

The observed 90%+ male dominance in chess grandmasters and Nobel science prizes exceeds these single-trait predictions. The multi-trait analysis in section 5 (see table) shows that applying GMVH across 3–4 independent relevant traits at the same threshold predicts 85–96% male fractions, which is much closer to the observed figures. Several additional factors further reconcile this:

1. Multiplicative trait effects: If GMVH applies independently to cognitive ability, risk-taking, competitive drive, and domain-specific interests, the combined effect of requiring extreme scores on multiple traits compounds the male-to-female ratio dramatically. A person must be in the top 0.01% of cognitive ability AND risk tolerance AND competitive drive AND relevant interests to reach chess grandmaster or Nobel laureate level.

2. Historical non-meritocratic exclusion: Women were historically excluded from many fields, creating pipeline deficits that persist today. Nobel prizes awarded before 1970 reflect near-total exclusion; this inflates the historical male percentage beyond what GMVH would predict.

3. Self-selection by extreme traits: At the extreme end of chess or research careers, the willingness to invest 10,000+ hours in a narrow discipline correlates with extreme trait combinations that GMVH would predict are rarer in women.

Verdict: 🟡 Partially Supported. GMVH predicts men will be overrepresented, and the direction is confirmed. The magnitude (especially in objectively scored domains) can be reconciled with GMVH through multiplicative effects across traits, but single-trait variance ratios alone do not fully account for observed 97%+ male dominance. Historical factors and compounding trait requirements also contribute.

Summary Table

Overall: Largely True — The core logical inference (more men at top positions under meritocracy if GMVH is true) is mathematically sound and empirically consistent across most industries. The multi-trait GMVH prediction (5–14% female at extreme combined thresholds, widening to ~5–25% under correlated-trait assumptions) aligns closely with observed figures of ~10% female Fortune 500 CEOs and ~7% female heads of government. The prediction of low but non-zero female representation at the top is confirmed. The only meaningful overstatement in the original claim is “all industries”: in fields where GMVH-relevant traits are not the primary success factors (healthcare, education, social work), women already dominate without any GMVH-predicted disadvantage.

References

Primary Sources

1. Baye A, Monseur C. “Gender differences in variability and extreme scores in an international context” Published: 2016 | Journal: Large-scale Assessments in Education 4(1):1 URL: https://link.springer.com/article/10.1186/s40536-015-0015-x Key finding: “The greater male variability hypothesis is confirmed” — boys show 14% greater variance across 12 international assessment databases (PISA, TIMSS, PIRLS, 1995–2015)

2. Volk S, Thöni C, Ruigrok W. “Converging evidence for greater male variability in time, risk, and social preferences” Published: 2021 | Journal: PNAS 118(23): e2026112118 URL: https://pmc.ncbi.nlm.nih.gov/articles/PMC8201935/ Key finding: Meta-analysis of 50,000+ individuals across 97 samples finds variance ratio of 1.25 for economic preferences — men have significantly more extreme preferences in time, risk, and social domains

3. Heterodox Academy. “The Greater Male Variability Hypothesis” Published: 2017 (updated) | Accessed: March 2026 URL: https://heterodoxacademy.org/blog/the-greater-male-variability-hypothesis/ Key finding: Comprehensive review concludes GMVH is real, consistent across decades, and has practical implications for hiring at elite levels

4. Fortune. “Women CEOs run 10.4% of Fortune 500 companies” Published: June 2024 | Accessed: March 2026 URL: https://fortune.com/2024/06/04/fortune-500-companies-women-ceos-2024/ Key finding: Only 52 of 500 Fortune 500 companies (10.4%) led by women as of 2024; first crossed 10% threshold in 2023 after 70 years

5. Pew Research Center. “About a third of UN member states have ever had a woman leader” Published: October 2024 (updated March 2026) | Accessed: March 2026 URL: https://www.pewresearch.org/short-reads/2024/10/03/women-leaders-around-the-world/ Key finding: Only 63 of 193 UN countries have ever had a woman head of government; 13 currently (~6.7%) led by women

6. Wikipedia. “Variability hypothesis” Accessed: March 2026 URL: https://en.wikipedia.org/wiki/Variability_hypothesis Key finding: Historical overview of GMVH, its confirmation in multiple studies, and key debates; Thorndike (1906) predicted male variability would mean “eminence and leadership of the world’s affairs will inevitably belong oftener to men”

7. Ahern KR, Dittmar AK. “The changing of the boards: The impact on firm valuation of mandated female board representation” Published: 2012 | Journal: Quarterly Journal of Economics 127(1): 137-197 URL: https://gap.hks.harvard.edu/changing-boards-impact-firm-valuation-mandated-female-board-representation Key finding: Norway’s 40% board gender quota led to less-experienced boards (new female directors had 31% CEO experience vs. 65% for male incumbents) and decline in firm value — consistent with a smaller pipeline of women with equivalent senior experience

8. FIDE (World Chess Federation) rating database Key finding: 1,643 male Grandmasters vs. 37 female Grandmasters — 97.8% male — in an entirely objective, merit-based ranking system

Evidence Files