ArgentinaThe Institutional Gap
Strong job outcomes and complete economic data coverage. Institutions remain the weakest dimension.
Women’s economic activity is visible in the data, yet access to work and the mechanisms that support economic rights still lag. Institutional visibility accounts for 44% of the gap to full parity.
Bars show the five reported dimension scores. Animation reveals the profile, not a change over time.
Abstract
Argentina scores 78.9 out of 100 on the Gender Economic Visibility Index (GEVI), in the Emerging band, with a 90% uncertainty interval of 74.4 to 81.7 and a point rank of 8 among 19 G20 economies. Job quality and statistical coverage are strong, but institutional visibility remains the principal constraint. All 13 inputs are observed: 12 are grade A, one is grade B and none is imputed (blind-spot ratio 0.00).
Women’s labour force participation is 19.2 percentage points below men’s. Among those employed, women have lower vulnerable-employment shares and higher wage-employment shares, but occupy only 36.4% of senior and middle management roles. Strong conditional job outcomes coexist with a gap in access to work.
D4 institutional visibility is the principal constraint at 0.500. Shapley attribution assigns it 9.25 of the 21.1 points missing from full parity. The largest grouped scenario halves labour-access parity gaps, raising the index to 81.3. These are conditional model results, not causal policy estimates.
Download full paper PDFInputs and parity calculations
All 13 inputs are observed. Twelve are grade A and one is grade B. Each value retains its original reference year and evidence grade.
| Indicator | Women | Men | Year | Score | Grade |
|---|---|---|---|---|---|
| I1 Labour force participation (%) | 52.5 | 71.7 | 2025 | 0.732 | A |
| I2 Youth NEET (%) | 18.1 | 14.7 | 2024 | 0.812 | A |
| I3 Youth unemployment (%) | 21.0 | 17.8 | 2025 | 0.848 | A |
| I4 Vulnerable employment (%) | 20.5 | 23.6 | 2025 | 1.000 | A |
| I5 Women in senior and middle management (%) | 36.4 | — | 2024 | 0.728 | A |
| I6 Wage and salaried employment (%) | 77.2 | 72.4 | 2025 | 1.000 | A |
| I7 Firms with female ownership participation (%) | 57.8 | — | 2017 | 1.000 | B |
| I8 WBL legal frameworks (0–100) | 80.8 | — | 2025 | 0.808 | A |
| I9 WBL enforcement perceptions (0–100) | 67 | — | 2026 | 0.670 | A |
| I10 WBL supportive frameworks (0–100) | 50 | — | 2026 | 0.500 | A |
| I11 Women ministers (%) | 25.0 | — | 2024 | 0.500 | A |
| I12 Women MPs (%) | 40.9 | — | 2026 | 0.818 | A |
| I13 Data availability (12 of 12 rows) | — | — | 2026 | 1.000 | A |
NEET means not in education, employment or training; WBL means Women, Business and the Law. I4 scores non-vulnerable employment, calculated from the vulnerable shares above. I4, I6 and I7 are capped at 1; this indicates benchmark attainment, not equality in pay, hours or control of firms. I7 is nine years old and requires updating.
Sources used in the paper: World Bank Gender Data Landscape, April 2026 edition (including WBL 2026), and IPU parliamentary ranking, 1 August 2026. This abstract summarises the supplied research paper dated 29 September 2026; the estimates are not official statistics.
Step 1 · Indicator calculations
Scores range from 0 to 1. Favourable outcomes use women/men; adverse outcomes use men/women. Women’s representation is benchmarked against 50%, WBL scores against 100, and coverage against 12 rows.
| Indicator | Calculation | Score | Grade |
|---|---|---|---|
| I1 | min(52.5 / 71.7, 1) | 0.732 | A |
| I2 | min(14.7 / 18.1, 1) | 0.812 | A |
| I3 | min(17.8 / 21.0, 1) | 0.848 | A |
| I4 | min((100 − 20.5)/(100 − 23.6), 1) | 1.000 | A |
| I5 | min(36.4 / 50, 1) | 0.728 | A |
| I6 | min(77.2 / 72.4, 1) | 1.000 | A |
| I7 | min(57.8 / 50, 1) | 1.000 | B |
| I8 | 80.8 / 100 | 0.808 | A |
| I9 | 67 / 100 | 0.670 | A |
| I10 | 50 / 100 | 0.500 | A |
| I11 | min(25.0 / 50, 1) | 0.500 | A |
| I12 | min(40.9 / 50, 1) | 0.818 | A |
| I13 | 12 / 12 | 1.000 | A |
Dimensions, aggregation and uncertainty
Each dimension is the simple mean of its indicators. The weighted geometric mean limits the extent to which strong dimensions can compensate for weak ones.
| Dimension | Mean of indicators | Score D | Weight w | Dʷ |
|---|---|---|---|---|
| D1 Labour participation | (I1 + I2 + I3)/3 | 0.797 | 0.25 | 0.9450 |
| D2 Job quality and seniority | (I4 + I5 + I6)/3 | 0.909 | 0.20 | 0.9812 |
| D3 Enterprise and legal rights | (I7 + I8 + I9)/3 | 0.826 | 0.20 | 0.9625 |
| D4 Institutional visibility | (I10 + I11)/2 | 0.500 | 0.15 | 0.9013 |
| D5 Statistical and political voice | (I12 + I13)/2 | 0.909 | 0.20 | 0.9811 |
= 78.9
Step 3 · Aggregation trace
Using unrounded inputs, the five weighted terms multiply to 0.78906. Equivalently, Σ(w ln D) = −0.23692 and exp(−0.23692) = 0.78906. Multiplying by 100 gives 78.91, reported as 78.9. Rank 1 is highest; Argentina ranks 8 of 19. D2 and D5 both round to 0.909, with D2 marginally higher at full precision.
Step 4 · Uncertainty and robustness
The source’s 10,000-draw joint Monte Carlo simulation samples indicator scores from normal distributions truncated to [0, 1] and dimension weights from a Dirichlet distribution centred on the base weights, with concentration 50 and seed 2026. Grade A scores use SD 0.02 and grade B SD 0.06; these are judgemental settings. The median GEVI is 78.4 and the 5th–95th percentile interval is 74.4–81.7. Rank has median 9 and a 5th–95th percentile range of 8–10.
Weight uncertainty accounts for about 78.2% of simulated output variance. Seven weighting schemes yield scores of 66.0–80.6 and ranks of 8–11. Benchmark and cap specifications yield 75.8–82.9 while retaining rank 8; imputation choices do not change Argentina’s score because no input is imputed. The interval crosses the Visible threshold of 80, so the Emerging classification should be read with uncertainty.
Where the gap is
Gain from lifting each dimension to full parity, holding the others fixed. These are separate counterfactuals and must not be added together.
| Dimension raised to 1.0 | GEVI | Gain | Shapley gap points |
|---|---|---|---|
| D1 | 83.5 | +4.6 | 5.04 |
| D2 | 80.4 | +1.5 | 1.69 |
| D3 | 82.0 | +3.1 | 3.41 |
| D4 | 87.6 | +8.6 | 9.25 |
| D5 | 80.4 | +1.5 | 1.70 |
Shapley attribution averages marginal gains across all 120 dimension orderings. It allocates the 21.1-point gap to parity without double counting: D4 contributes 9.25 points (43.9%), followed by D1 at 5.04 points (23.9%).
Selected indicator and grouped scenarios reported in the source:
| Scenario | GEVI | Gain |
|---|---|---|
| I10 supportive frameworks: 0.50 → 0.60 | 80.0 | +1.14 |
| I11 ministerial parity: 0.50 → 0.60 | 80.0 | +1.14 |
| I12 parliamentary parity: 0.818 → 0.918 | 79.8 | +0.85 |
| S3 Legal and enforcement scores each +0.10 | 80.1 | +1.2 |
| S4 Halve parity gaps in I1, I2, I3, I4 and I6 | 81.3 | +2.4 |
| S7 Combined legal, labour and enterprise scenarios | 82.7 | +3.8 |
S4 is the largest single grouped lever. In the greedy pathway, one +0.10 step in I10 reaches 80.0. These changes describe index arithmetic; they do not establish policy effects or predict a date of improvement.
Reading the result
Women’s labour force participation is 19.2 percentage points below men’s (52.5% against 71.7%), while youth unemployment is 3.2 points higher (21.0% against 17.8%). Among those employed, women have lower vulnerable-employment shares and higher wage-employment shares than men, but occupy only 36.4% of senior and middle management roles. Strong conditional job outcomes therefore coexist with a substantial gap in access to work.
Institutional visibility is the weakest dimension (D4 = 0.500): supportive frameworks score 50/100 and women hold 25% of ministerial posts. Legal frameworks score 80.8/100, compared with enforcement perceptions of 67/100 and supportive frameworks of 50/100. The results support examining the mechanisms needed to implement rights. Full coverage of the 12 economic data rows does not close the fiscal gap: the index contains no direct measure of women’s visibility in taxpayer records.
Priorities and practical use
Connect each priority to a named data product, owner and decision. Complete economic coverage does not remove the missing fiscal indicator.
Fund supporting mechanisms. Cost the gaps in equal-opportunity institutions, gender data units, budget tagging and services, then track I10 annually.
Make fiscal participation measurable. Add a sex field to wage-tax returns and taxpayer registrations, and publish an annual gender tax profile covering filing, withholding, refunds and credits. A proposed fiscal indicator would extend the model; it is not part of the current 78.9 score.
Measure implementation and representation. Publish complaints, prosecutions and remedies alongside appointments by sex. I11 reflects political appointments, which a revenue authority cannot directly change.
Refresh the evidence. Update the 2017 firm-ownership measure (I7), retain the complete 12-row coverage and re-run the index using comparable definitions and reference years.
Who uses the data, and for what decision?
| Data product | Primary user | Decision it informs | Cadence |
|---|---|---|---|
| I1–I3 labour access and youth outcomes by sex | Labour ministry, Treasury | Where to target participation and youth employment support | Annual |
| Costed inventory of missing support mechanisms | Treasury, gender ministry | Which institutional gaps to fund first | Budget cycle |
| Taxpayer and wage-tax statistics by sex | Revenue authority, Treasury | Who enters the formal tax base and receives refunds or credits | Annual |
| Ownership and core economic input ledger | Statistics office | Which dated inputs to refresh and how to preserve comparability | Survey plan |
| Enforcement and appointments series by sex | Justice and labour authorities, Cabinet office | Where implementation and representation gaps persist | Annual |
Read the score, interval and dimension profile together. The inputs span 2017–2026, the weights and uncertainty settings are assumptions, and the study is cross-sectional. Annual changes require comparable data and an assessment of change uncertainty; crossing 80 in a scenario alone is not evidence of a measured improvement.