This paper evaluates the heterogeneous effect of health insurance on out-of-pocket healthcare expenditure (OOPHE), using merged data from the Ghana Living Standards Survey and Ghana Health Service reports. It applies conditional-mixed process and censored quantile instrumental variable estimators to tackle censoring and endogeneity. We instrument household insurance rate with community insurance rate (exclusive of the observed household) and control for community unobservables. The quantile regression allows the insurance effect to differ across the distribution of OOPHE. We further perform separate analyses by the types of OOPHE and selected covariates. The results show that insurance reduced OOPHE and the incidence of catastrophic OOPHE in 2013, but not in 2017. Besides, households in the higher expenditure quantiles benefitted more from coverage than those in the median and lower quantiles did in 2013. Also, the insurance benefits accrued exclusively to the wealthiest households, households with older heads, and users of outpatient services in 2013. Lastly, the 2013 survey reveals that families with female heads and those whose heads had primary education benefitted more from health insurance than their counterparts in the other subgroups did. The paper concludes that same health insurance can have varied financial risk implications at different periods, across the distribution of OOPHE, and among various household categories.
Download the paper in PDF format
Samuel Ampaw, Simon Appleton and Xuyan Lou
View all CREDIT discussion papers | View all School of Economics featured discussion papers
Sir Clive Granger BuildingUniversity of Nottingham University Park Nottingham, NG7 2RD
Enquiries: hilary.hughes@nottingham.ac.uk