Health shocks, job quality, and self-employment in Africa

Abstract

This paper examines the relationship between occupation and different types of human capital—skills, education, ability and health. Summary statistics of our panel data from the Ghana Household Worker Survey strongly suggest that height and health vary by occupation. Our regression results show that after controlling for age, gender and movement out of jobs, the self-employed are the most likely to have at least one day of illness (a health shock) in the past year. However, conditional on having at least one day of illness, the self-employed have the lowest expected number of days ill. On the other hand, evidence that large firm workers have longer illnesses than other workers, perhaps reflecting their better employment circumstances that allow more time off when sick, cautions against the use of days of illness as a measure of health that is unbiased by occupational choice. We also investigate labour market outcomes and find that the number of days ill does not affect the labour supply decision. Controlling for entry into the labour force, however, days of illness does have a significant negative impact on earnings, especially if it exceeds 30 days. Our analysis suggests that changes in health and changes in occupation are strongly correlated; understanding the causal links in this relationship should be the focus of future work.