Abstract
This study investigates the relationship between public health expenditure and unemployment across 22 countries, comprising 11 developing and 11 developed countries, over the period 2003-2022. The dynamic panel threshold model is employed to explore non-linear effects of health spending on unemployment, using population density, control of corruption, and GDP per capita as threshold variables. A fixed effects model with interaction terms was further employed to assess robustness and test the moderating roles of conditional factors on the health expenditure and unemployment nexus.
The estimated results reveal that health spending has a significant impact on reducing unemployment among all three threshold variables in both developing and developed economies. For the population density threshold, the effects of health spending were significant and negative; however, in both countries, when the regime was below this threshold, it revealed larger unemployment-reducing effects. For the control of corruption threshold, both developing and developed economies have significant results above the threshold value, indicating that better governance has a larger effect on health spending, impacting unemployment. For the GDP per capita threshold, when health spending occurs in this regime, unemployment-reducing effects are significant. However, the magnitude of unemployment-reducing effects is mostly larger in developing countries. This suggests that health investment in developing countries is met with stronger employment multipliers due to the unmet healthcare needs, higher labour intensity, and spillover effects on supporting sectors. In contrast, the smaller marginal effects of health expenditure in developed countries reflect the diminishing returns to additional investment in health systems that are already strong and technologically advanced. Moreover, the threshold analysis reveals that beyond certain levels of population density, governance quality, and income levels, health expenditure has a weak effect on reducing unemployment.
The robustness estimation confirms these findings, as the interaction between health spending and the threshold variables remains significant and negative. This supports the context-specific nature of health spending, reducing unemployment. The study concludes that effective health investment can serve as a catalyst for employment opportunities, especially in developing economies. These findings contribute to the broader understanding of how health investment is able to influence labour market outcomes in heterogeneous economic contexts.