Spatial poverty and inequality in South Africa: A municipality level analysis

Abstract

Using the 2011 South African population census, we provide income and multidimensional poverty and inequality estimates at the municipal level. We go on to estimate a spatial econometric model to identify the correlates of poverty across municipalities in South Africa. Our results show that both income and multidimensional poverty and inequality vary significantly across municipalities in South Africa. In general, areas that are historically characterized by low economic and welfare outcomes still experience significantly higher poverty and deprivation levels. Using both global and local spatial autocorrelation measures we find significant and positive spatial dependence and clustering of regional development indicators. The situation of poverty is both spatially unequal and autocorrelated.

Results from our spatial econometric analysis indicate negative and significant relations between the municipal poverty levels and local levels of education and economic activity (GDP per capita). Significant and positive relations are found between municipal poverty levels and local inequality levels, suggesting that municipalities with higher levels of inequality also have higher incidences of poverty. In contrast, natural geographic factors such as rainfall and temperature are not significantly related to municipal poverty. Accounting for both direct, intra-municipality effects as well as spillover effects of neighbouring municipalities is important. These spillover effects notably reduce the coefficient sizes suggested by non-spatial, OLS regressions. Most striking, the large negative coefficient that OLS attributes to residing within a historical homeland area is greatly reduced and even loses statistical significance in some spatial models. Clearly municipalities in homeland areas are particularly likely to be surrounded by very poor municipal neighbours and therefore subject to strong negative spillovers. That said, when interactions between this historical geographical variable and contemporary socio-economic deprivations are included, then homeland becomes statistically significant once more. This makes the important point that while, it is these socio-economic deprivations that are particularly important in explaining contemporary income poverty across the county, those who reside in these homeland areas remain especially badly off in terms of these deprivations.