Determinants of tax revenue performance in the Southern African Development Community (SADC)

This paper investigates the determinants of tax revenue performance in all 15 Southern African Development Community countries during 1990-2010, using panel data. The investigation makes use of two estimation techniques in testing for country specificity. These are the least squares dummy variables fixed effects and the feasible generalised least squares by Park (1967) and Kmenta (1986). The extreme-bound analysis technique is also used in delineating the various causal relationships (including a sensitivity analysis). Prior to the estimation process, the study tested and controlled for applicable errors in the panel such as endogeneity, serial correlation, cross-sectional dependence of the error term, group-wise heteroscedasticity and contemporaneous correlation. The process addressed some major critique of panel data estimations involving large and small economies in a regional grouping like the SADC. The paper also introduces a value added tax harmonisation variable (and additionally made use of the corporate income tax harmonisation variables) through a tax policy harmonisation measure in investigating the impact of foreign direct investment and taxation on tax revenue collected. The results generally highlight the robust role of taxation (tax rates and tax policy harmonisation variables) (alongside other important determinants) in improving tax revenue in the region, providing empirical support for extant anecdotal evidence. The final empirical findings also confirm the importance of FDI inflows towards tax revenue collected in the SADC and the existence of reverse causality (that is, a causal relationship between FDI and taxation or vice versa). Policy considerations include the need for SADC countries to carry out extensive pro-regional (coordinated) tax reforms, create a regional tax forum and promote initiatives aimed at improving FDI and ultimately tax revenue (as per existing regional protocols).