Abstract

This study examines the heterogeneous impact of unemployment on health (i.e., life expectancy and under-five mortality) using unbalanced panel data for 48 countries in sub-Saharan Africa (SSA) from 2000 to 2020. We applied Generalized Quantile Regression in an instrumental framework with multi-criteria MCMC (Markov Chain Monte Carlo) optimization to address the endogeneity (simultaneous bias, omitted variable bias) between unemployment and health. The quantile estimation findings show that the impact of unemployment is heterogeneous (i.e., non-linear/asymmetric) across health outcomes distributions. Unemployment significantly reduces life expectancy in the 10th, 20th, 30th, 50th, 60th, 80th, and 90th quantiles while increasing under-five mortality in the 10th, 50th, 60th, and 70th quantiles. The possible mechanism influencing the unemployed to have negative health effects could be the role of poverty associated with a lack of income. Meanwhile, the life expectancy of the unemployed in the 40th quantile is increasing significantly, which may be linked to the fact that more physical activity has a positive effect on their health. Similarly, unemployment significantly reduces under-five mortality in the 20th, 40th, 80th, and 90th quantiles, and the possible mechanism for the unemployed to reduce under-five mortality may be through financial support from family members who can help them buy nutritious food and provide medical care for children. Bayesian robustness tests support evidence prior to the COVID-19 outbreak that unemployment increases under-five mortality and creates uncertainty about life expectancy from 2000 to 2019. Overall, the findings for the period 2000–2020 confirm that unemployment has asymmetric (positive and negative) effects on health in SSA. The findings suggest two conclusions. First, unemployment is linked to poor health (lower life expectancy) and higher mortality in sub-Saharan Africa. Second, not all unemployment in the region is unhealthy. The policy implications for SSA countries are outlined.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call