Peer-reviewed article
Disentangling wealth and stunting: structural and macroeconomic determinants of child nutrition across countries
Amr Abdelkader Saber ·
Summary
Written by AI, not yet AICS's view · claude-sonnet-5-5This study asks whether the link between national income and child stunting reflects short-term effects of income or long-standing structural differences between countries. It analyses data from 104 countries over 2000 to 2020 using several panel regression methods. Almost all of the income–stunting association came from differences between countries, not from changes within a country over time. Short-run inflation showed no robust link to stunting. The authors conclude that long-term investments matter more for stunting than short-term economic swings.
Key points
- A balanced panel of 104 countries from 2000 to 2020 was analysed.
- About 95% to 96% of the income–stunting association was due to cross-sectional differences between countries rather than short-run within-country changes.
- Pooled OLS linked a 1% rise in GDP per capita to a 0.84 percentage-point reduction in stunting prevalence.
- After controlling for country and year effects, the within-country effect fell to 0.03 percentage points.
- No statistically robust short-run within-country association was found between consumer price inflation and stunting.
Limitations
This is a country-level analysis, so it cannot show what happens for individual children or households, and it does not name the countries or test a specific programme. The abstract does not give details of data quality, and the policy recommendations are conclusions drawn by the authors rather than tested interventions.
Abstract
Background: This study investigates whether the observed negative relationship between national income and child stunting prevalence reflects short-term causal effects or long-term structural differences between countries. Methods: A balanced panel of 104 countries from 2000 to 2020 was analyzed using pooled ordinary least squares and two-way fixed effects estimators. Robustness checks included lagged specifications, Hausman tests, and dynamic panel System Generalized Method of Moments models with clustered standard errors. Results: Approximately 95% to 96% of the observed income–stunting association was attributable to cross-sectional differences rather than short-run within-country dynamics. While pooled ordinary least squares suggested that a 1% increase in gross domestic product per capita was associated with a 0.84 percentage-point reduction in stunting prevalence, the within-country effect declined to 0.03 percentage points after controlling for country and year effects. No statistically robust short-run within-country association was detected between aggregate consumer price index inflation and stunting prevalence. Conclusions: The findings suggest that child stunting is shaped more by long-term structural determinants than by short-run macroeconomic fluctuations. Policies should prioritize sustained investments in institutional quality, health systems, sanitation, maternal education, human capital, and social protection.