Social Protection, Health Risk, and Household Welfare in Zambia
Abstract
Households in sub-Saharan Africa face substantial health risk. This threatens their welfare and predisposes them to poverty. Despite the high risk environment, they have little or no access to social protection–a set of programs that aims to reduce health risk and provides insurance against its effects, key of which are reductions in labor income and increases in household health expenditure. In childhood, health risk may have additional effects; it lowers cognitive abilities as well as educational attainment and these effects persist in adulthood, working to permanently lower lifetime economic outcomes. Yet still, children from poorer backgrounds face a disproportionately larger share of childhood health risk. In this thesis, I examined the extent to which households are protected from the welfare effects of health shocks (illness and injury) in Zambia. I also evaluated some social protection policies focused at the general population and specific groups such as children and individuals from low socioeconomic background. This was achieved in three sub-studies, each of which forms a separate paper. The first one assessed the effect of health shocks on household consumption, income, and health spending, as well as the extent to which households use borrowing and selling assets as coping strategies in the absence of complete social protection systems, during and after structural adjustment reforms (SAPs). Using data from four waves of the living conditions monitoring survey (LCMS) in the period 1996–2006, it was found that health shocks were associated with reduced consumption both during and after structural reforms. Although health shocks were substantially associated with reduced labor income in both periods, the effect on health spending was much greater after the structural reforms. Middle income households were especially vulnerable. To cope with this risk, household employed informal borrowing and selling assets as self insurance mechanisms. In the second paper, the short and long term effects of an important social protection policy–the user fee removal–on medical spending and overall utilization of health services was evaluated. Heterogeneity in utilization response was also examined. Results show that the policy increased overall utilization of health services in the short term and these effects were sustained in the long term. Apart from increasing overall utilization, the policy also led to shifting of use from private to public health services. The greatest increase in utilization of health services occurred among individuals whose household heads were either unemployed or had no education. Further, although the policy reduced the proportion of individuals incurring any spending, overall health expenditure was not affected in any significant way. Third, and finally, the last paper investigated the determinants of childhood health risk, specifically stunting and fever, between 2007–2014, a period of massive scale up of child health interventions as countries braced themselves to meet the 2015 target of the Millennium Development Goal on child health. It assessed whether or not the concentration of health risk among children from poorer households reduced. Importantly, the factors or determinants that could have been driving these changes were investigated. It was found that although the prevalence of stunting in the general population and in all quartiles, except the poorest, reduced, inequality increased significantly. The determinants that contributed the most to the increase in inequality of stunting were maternal height and weight, household wealth, birth order, place of birth (home or facility), breastfeeding duration and maternal education. Socioeconomic inequality in fever also increased and incidence of fever did not fall. The determinants that contributed to the increase in the inequality of fever were household wealth, maternal education, birth order, and duration of breast feeding. I conclude that scaling up social protection programs that aim at providing insurance against health risk would improve household welfare, especially if coverage does not only focus on the poorest but also middle income households, who are found to be most vulnerable. Regarding user fee removal, although this policy may reduce health risk, since it increased utilization of health services, especially among individuals from low socioeconomic backgrounds, it was not successful in reducing health expenditure risk. Other social protection programs need to be considered if there has to be improvement in health spending insurance. In the same vein, childhood health risk became more concentrated among children from poorer households despite the massive scale up in child health interventions. Reducing inequality in the determinants of childhood health such as facility deliveries, wealth, education, nutrition, etc is key to reducing inequalities in childhood health risk. If inequalities of determinants are not eliminated, increasing their coverage may not reduce child health inequality, and may, in fact, increase it.
Has parts
Paper I: Peter Hangoma, Arild Aakvik, and Bjarne Robberstad. Health shocks and household welfare in Zambia: An assessment of changing risk. Full text not available in BORA.Paper II: Hangoma, Peter, Bjarne Robberstad, and Arild Aakvik. "Does Free Public Health Care Increase Utilization and Reduce Spending? Heterogeneity and Long-Term Effects." World Development (2017). The article is available at: http://hdl.handle.net/1956/16646
Paper III: Hangoma, Peter, Arild Aakvik, and Bjarne Robberstad. "Explaining changes in child health inequality in the run up to the 2015 Millennium Development Goals (MDGs): The case of Zambia." PloS one 12.2 (2017):e0170995. The article is available at: http://hdl.handle.net/1956/16648