How many children, men, and women are living in poverty globally?

An approach combining structural and machine-learning models

2026-08-07

Disclaimer and Acknowledgements

The findings, interpretations, and conclusions expressed in this presentation are entirely mine. They do not necessarily represent the views of the World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. I gratefully acknowledge financial support from the UK government through the Data and Evidence for Tackling Extreme Poverty (DEEP) Research Programme and the Umbrella Facility for Gender Equality (UFGE) of the World Bank.

Introduction

Motivation

Standard per-capita measures of global poverty assume:

  1. Equal distribution of household resources
  2. All consumption is private
  3. Identical needs across individuals

Motivation

Standard per-capita measures of global poverty assume:

  1. Equal distribution of household resources
  2. All consumption is private
  3. Identical needs across individuals

But the evidence shows:

  1. Inequality within households (men vs. women vs. children)
  2. Sharing of resources (e.g., housing) thus economies of scale
  3. Different needs (adults vs. children)

This is the first global study to integrate all three facts.

  • Identifies poor individuals (not just poor households)
  • Improves poverty monitoring and policy targeting

Key findings

  • Of the 805 million extreme poor in 2024:
    • Women: 40%
    • Men: 30%
    • Children: 30%
    • Children in Sub-Saharan Africa: 24%
    • Women in Sub-Saharan Africa: 20%
  • Children and women especially in Sub-Saharan Africa are disproportionately poor.
  • About quarter of a million children are poor.
  • 82 million more women than men are poor.

Millions living in extreme poverty, 2024

Presentation outline

  • How do we incorporate the three facts into poverty measurement?
    • Structural models, Machine learning, Derive global distribution (inequality)
    • Derive a scaled, adult-equivalized poverty line (sharing, different needs)
  • Global and regional poverty estimates
  • Conclusion

Empirical strategy of this paper

Combine:

  • structural estimates of intra-household resource shares for 45 countries

with

  • machine-learning predictions for the remaining countries

to construct an individualized global consumption distribution and re-estimate poverty.

Country group Women /Men Children /Adult Obs Pop. (%)
East Asia & Pacific 0.70 0.14 2 0.2
Europe & Central Asia 0.90 0.24 5 3
Latin America & Caribbean 0.88 0.45 11 77
Middle East & North Africa 0.72 0.20 4 42
South Asia 0.86 0.38 2 97
Sub-Saharan Africa 0.78 0.26 21 70
Low-income countries 0.81 0.26 15 61
Lower-middle-income 0.83 0.33 15 75
Upper-middle-income 0.84 0.40 13 21
High-income countries 1.03 1.03 2 2
All 0.83 0.34 45 42

Source: Aminjonov et al. (2026)

Ratio of women-to-men resource shares from Aminjonov et al.

Machine-learning models

Outcome variables from Aminjonov et al.

Prediction variables from different data sources

Source Description Countries Period Variables
World Bank Gender Data Portal 217 1960–2025 1383
Gethin & Saez (2025) Hours worked by gender/age 159 1900–2023 189
Gallup Polls Cultural values and beliefs 165 2006–2024 81
World Bank Poverty & Inequality Platform 218 1977–2026 28
Pew Research Center Population shares of religions 195 1990–2024 15
World Values Survey Cultural values and beliefs 65 2000–2024 13
United Nations Gender Development Index 195 1990–2023 11
World Bank World Governance Indicators 208 1996–2024 6
World Economic Forum Global Gender Gap Index 160 2006–2025 5
Jolliffe et al. (2025) Food budget share 167 1990–2024 1
World Inequality Lab Female labor income share 211 1990–2023 1
All data 218 1900–2026 1733

XGBoost is the preferred ML model

  • Linear models: LASSO, Ridge, Elastic Net
  • Non-linear models: Random Forest, XGBoost

Why XGBoost?

  • Model builds trees sequentially
  • Each tree corrects previous errors

\[ \hat{y}_i^{(t)} = \hat{y}_i^{(t-1)} + f_t(x_i) \]

  • Captures interactions
  • Handles missing data
  • Regularization controls risk of overfitting (p=1733 ≫ n=45)

Feature trimming, hyperparameter tuning, model selection

  • Staged grid search:
    • Stage 1: predictors, tree depth, minimum child weight
    • Stage 2: learning rate, boosting rounds
    • Stage 3: λ, α, γ, subsample, column sample
  • Model selection (248 specifications):
    • Leave-One-Out CV (LOOCV)
    • Optimize out-of-sample R²
  • Key results:
    • Inequality by gender: R² ≈ 0.74
    • Inequality by age: R² ≈ 0.78
Top predictors of women-to-men resource share
Variable (4/15) Gain
Female GNI per capita (2021 PPP$) 0.229
Women’s employment rate 0.123
Male bachelor’s degree attainment (25+) 0.097
Female-to-male ratio of hours worked, elderly 0.092
Top predictors of child-to-adult resource share
Variable (4/10) Gain
GDP per capita (current US$) 0.263
Female employment in services (%) 0.184
Life expectancy at age 60, female (years) 0.176
Female youth illiteracy rate (%) 0.133

Note: Gain reflects each variable’s contribution to reducing prediction error.

Ratio of women’s share to men’s share - structural model

Ratio of women’s share to men’s share - structural model + ML model

Ratio of children’s share to adults’ share - structural model + ML model

Individual-level global distribution and scaled poverty line

From Ratios to Resource Shares

  • Step 1: Define inequality ratios

\[ \frac{w_{shr}}{m_{shr}} = wm_{gap} \qquad\qquad \frac{c_{shr}}{0.5(m_{shr} + w_{shr})} = ca_{gap} \tag{1–2} \]

  • Step 2: Resource constraint, total private household expenditure/resources:

\[ T_{prv} = M \cdot m_{shr} + W \cdot w_{shr} + C \cdot c_{shr} \tag{3} \]

  • Step 3: Solve system (3 equations, 3 unknowns)

\[ m_{shr} = \frac{T_{prv}} {M + wm_{gap}\cdot W + 0.5\cdot ca_{gap}\cdot(1+wm_{gap})\cdot C} \tag{4} \]

\[ w_{shr} = wm_{gap}\cdot m_{shr} \qquad c_{shr} = 0.5\cdot ca_{gap}\cdot(1+wm_{gap})\cdot m_{shr} \tag{5-6} \]

Welfare Aggregation and Equivalent Poverty Line

Total household welfare is decomposed as:

\[ T = T_{pub} + T_{prv} \tag{7} \]

\[ T_{pub} = welf_{pc}\,(1 - pvt)\,N \tag{8} \]

\[ T_{prv} = Nwelf_{pc}\,pvt\,N/A \tag{9} \]

Note:
- Household size, \(N = W + M + C\)
- Adult-equivalent size, \(A = W + M + 0.6C\)
- \(welf_{pc}\): welfare per capita (2021 PPP$/day)
- \(pvt\): private share of household welfare

Allocation:
- Public consumption enjoyed by everyone
- Private consumption allocated via Eqs. (1-3)

Adjustments account for:
- Economies of scale
- Different needs of children and adults

Scaled, adult-equivalized poverty line \(z\) solves:

\[ F(z) = \int_0^z f(y(x))\,dx = P^0 \tag{10} \]

where:
- \(f(y(\cdot))\): transformed global distribution
- \(z\): equivalent poverty line
- \(P^0\): global per-capita poverty rate (12%)

Poverty lines used:
- International poverty line: $3.00 per capita
- Equivalent international poverty line: $7.70 per adult [60% for children (i.e. under 14)]

Global and regional poverty estimates

Extreme poverty rate by gender, age, region - per capita approach

  • Global poverty rates, 2024
    • Women: 9%
    • Men: 8%
    • Children: 21%
    • All: 12%
  • Men and women have similar poverty rates.
  • Children are much poorer than adults.

Extreme poverty rate by gender, age, region - alternative approach

  • Global poverty rates, 2024
    • Women: 12%
    • Men: 9%
    • Children: 14%
    • All: 12%
  • Child poverty falls by 7pp.
  • Still, children are poorer than adults (Africa effect).
  • Women face higher poverty risk than men.

Millions of extreme poor by gender, age, region - alternative approach

  • Of the 805 million extreme poor in 2024:
    • Women: 40%
    • Men: 30%
    • Children: 30%
    • Children in Sub-Saharan Africa: 24%
    • Women in Sub-Saharan Africa: 20%
  • Children and women especially in Sub-Saharan Africa are disproportionately poor.
  • About a quarter of a million children are poor.
  • 82 million more women than men are poor.

Conclusion

Global poverty looks different once we move from poor households to poor individuals.

The paper’s central conclusion is:

  • women are substantially more likely than men to live in extreme poverty
  • child poverty is the highest, even after accounting for sharing, inequality, and needs
  • the demographic and regional composition of global poverty changes materially once we account for sharing, inequality, and needs

Presented six times in a year

  • Atkinson Conference on Economic and Social Inequality
    University of Oxford, Oxford, UK — September 2026

  • European Development Economics Group (EDEG)
    Kiel, Germany — June 2026

  • Conference on Economic Measurement 2026
    London, UK — May 2026

  • 23rd Midwest International Economic Development Conference (MWIEDC)
    Madison, Wisconsin, USA — April 2026

  • Research Retreat, Chair of Development Economics
    ETH Zurich, Switzerland — January 2026

  • International Inequalities Institute (III) 10th Anniversary Conference
    LSE, London, UK — September 2025

Thank you

Comments welcome

Samuel Tetteh-Baah
World Bank

stettehbaah@worldbank.org

Important limitations

  • structural estimates are available for only 45 countries
  • machine-learning prediction is informative, but imperfect
  • country-level inequality parameters are applied uniformly within countries
  • the current version focuses on 2024, not trends

Still

Even with these constraints, the paper provides a meaningful first global picture of poverty by age and gender under more realistic household assumptions.