Local government health expenditure and life expectancy: cross-regency evidence from West Kalimantan, Indonesia

Abstract

While subnational health spending in Indonesia has expanded under fiscal decentralisation, stark disparities in life expectancy persist across regions, raising questions about whether larger budgets translate into better population health. Grounded in the health production function and capability approach, this study examines the effect of local government health expenditure on life expectancy at birth across all 14 regencies and municipalities in West Kalimantan Province and tests whether this relationship holds after accounting for regional poverty. Using 2023 cross-sectional data from Statistics Indonesia (BPS) and the Directorate General of Fiscal Balance (DJPK), we estimate log-linear regression models via Ordinary Least Squares (OLS) with MacKinnon-White-Davidson (HC3) heteroskedasticity-robust standard errors, complemented by outlier sensitivity checks. In the baseline specification, health expenditure exhibits a positive and statistically significant association with life expectancy (; HC3 ), explaining 53.1% of cross-regional variation; a 10% increase in spending corresponds to approximately 0.28 additional years of life expectancy. However, once the regional poverty rate is controlled for, the health expenditure coefficient drops by 72% () and loses statistical significance (; HC3 ), whereas poverty exerts a strong negative effect () and raises the model's explanatory power to 89.3%. These findings demonstrate that nominal budget size is not an autonomous determinant of longevity when socioeconomic deprivation constrains households' capability to access and benefit from public health services. Local governments should shift from input-based budgeting toward performance-based allocations targeting primary and promotive-preventive care in high-poverty inland regencies, supported by needs-based provincial fiscal transfers.