The determinants of government expenditure on education and health in SADC countries: a panel data approach
Government expenditure on education and health are considered to be significant for SADC development and growth. Majority of economies in SADC region experience extreme poverty which is less than US$1.25 per day according to the international standard (UN 2017). Health system in the region is the poorest in the world where children and adults die from diseases which can be easily cured such as malaria, TB, HIV and so forth. Furthermore, education in the region is not widely provided to everyone who deserve it (UN, 2017). All this challenges stated above transpire while the economies in SADC spend larger portions of their national budget on both education and health. Therefore, it is crucial to examine the factors that are responsible of their variation. This study explores the essential determinants of education and health expenditure using annual data covering the period 1997 to 2016 for SADC countries. Which include Angola, Botswana, and Democratic Republic of Congo, Lesotho, Madagascar, Malawi, Mauritius, Mozambique, Namibia, Seychelles, South Africa, Swaziland, Tanzania, Zambia, and Zimbabwe This study applies Levin, Lin and Chu (2002), and Im, Pesaran and Shin (2003) to test for unit panel root in the series. These two test are most recent tests in panel unit root test, with Levin et al (2002) assuming a common unit root process, and Im et al (2003) assuming individual unit root process, that is first order autoregressive parameters differ among the cross section. All variables were stationary at level except for education expenditure, health expenditure, life expectancy, and tertiary enrolment. Panel cointegration is applied using Pedroni, and Kao cointegration tests. In both Pedroni, and Kao cointegration confirms cointegration among the variables. FMOLS and DOLS are applied to estimate the parameters of each variable understudy. Granger causality is also applied to test for causality between the dependents and independent variables. It was established that both the methods of Dynamic OLS and Fully Modified OLS were consistent on most variables. On education expenditure model, both DOLS and FMOLS showed there is positive and significant relationship between health, population and education expenditure. Furthermore, on education expenditure both DOLS and FMOLS showed that there is negative and significant relationship between economic growth, corruption and education expenditure. However, tertiary enrolment results were not consistence between the estimators. DOLS showed negative and insignificant relationship between tertiary enrolment and education expenditure, whereas, FMOLS showed the opposite. On health expenditure model, life expectancy was found to be positive and insignificant related with health expenditure for both FMOLS and DOLS. Education expenditure was found to be positive and significantly related with health expenditure. Economic growth was found to be negative and significant related with health expenditure under FMOLS, while the opposite was found under DOLS. Population growth was found to be positive and significant relation with health expenditure under FMOLS, while is insignificant related under DOLS. Corruption was also found to be positive and negative, and significant related to health expenditure under FMOLS and DOLS respectively. On education expenditure, Granger causality test revealed that there is one way causality running from corruption to education expenditure at 1% level of significant. Furthermore, causality was found running from population growth to government expenditure on education at 1% level of significant, in SADC countries. On health model, causality was found running from education expenditure to health expenditure at 1% level of significant. On health expenditure, this study recommends for SADC governments to implement policies that increase education expenditure and improve measures which reduce corruption in addressing health problems. On education expenditure, this study recommends that the government of SADC implement policies that are pro economic growth, and measures that discourages corruption and population growth control policies to improve education expenditure.