The Effect of Household Heads’ Health on Spousal Labour Supply Decisions in Ghanaian Households

Abstract

This study examines the effect of household heads’ health on spousal labour supply in Ghana, using data from the Ghana Annual Household Income and Expenditure Survey (AHIES). Existing studies on health and spousal labour supply have largely been conducted outside Ghana and have mainly emphasized gender dynamics. This study extends the literature by investigating whether the health of the household head influences the labour supply of the unaffected spouse across male- and female-headed households, thereby providing new evidence on how intra-household roles shape labour supply decisions in the presence of health disruptions. The study also examines the buffering role of health insurance in shaping spousal labour supply responses when household heads are ill or injured. Spousal labour supply is measured as weekly hours worked. The distribution of hours worked is discrete and right-skewed, making count data models more appropriate than linear regressions. The main estimation strategy is a random-effects negative binomial model, which accommodates over-dispersed count outcomes. For robustness, a fixed-effects Poisson model with robust standard errors is also estimated. A severity-of-ill-health variable was derived from the health measure to capture three categories of household heads' health: healthy, unhealthy but continuing usual activities, and unhealthy and stopping usual activities. The results reveal gendered differences in how spousal labour supply responds to household heads’ health. Among male-headed households, female spouses significantly reduce their working hours when the household head is in poor health, consistent with increased caregiving responsibilities. In contrast, among female-headed households, male spouses’ labour supply remains largely unchanged.