The International Monetary Fund and Neonatal Mortality Rates, 1985-2018

Background
In 2015, 193 nations committed to UN Sustainable Development Goal 3·2, which aims to reduce child mortality rates by 2030. Despite progress, challenges remain, especially in low- and middle-income countries. International Monetary Fund (IMF) lending programs may influence these outcomes through structural adjustments which prioritise macroeconomic stability over public spending. In this study, we examine the impact of IMF interventions on neonatal mortality, addressing concerns about existing analyzes with a novel statistical approach.
 
Methods
We used neonatal mortality data from 1985-2018 across 106 countries, controlling for economic, political, and social factors. Our analysis employed a "stacked regression" approach, focusing on first-time entries into IMF programs to avoid biases from repeated program entries and exits. We also assessed dose-response relationships between IMF program stringency and neonatal mortality. Finally, we explored possible causal pathways and conducted tests of robustness using alternative models and definitions.
 
Findings
First-time participation in an IMF program was associated with a 5·8% increase in neonatal mortality rates, a finding that was robust across different model specifications and control group definitions. We also found a dose-response relationship, with each additional IMF condition increasing neonatal mortality by 0·032% on average. IMF program were linked to declines in GDP per capita, government expenditure, and healthcare workforce size, suggesting these as possible causal pathways. Our findings are consistent with earlier studies on both child mortality outcomes and possible mechanisms linking them to IMF programs, which we summarize in the online appendix.