Poverty in South Africa varies greatly across the nine provinces. An accurate estimation of relative poverty shares is important because they serve as key indices for targeting social expenditure. In this article we test the robustness of provincial poverty rankings against changes in measurement methodology. In recent years, a large body of international literature has developed concerning the choice of an appropriate poverty line and the construction of more appropriate poverty measures. This article uses two of these recent developments ‐ the concept of a poverty critical range in place of a single poverty line and distribution‐sensitive decomposable poverty measures — to re‐examine provincial poverty. Results are checked across two recent national data sets.
Notes
Respectively Senior Lecturer, School of Economics, University of Cape Town; Lecturer, Department of Economics and Economic History, University of Port Elizabeth. This is a revised version of a paper read at the Conference of the Economic Society of South Africa held at Potchefstroom University for Christian Higher Education, 8‐9 September 1997. The work of Murray Leibbrandt was supported by a research grant from the University of Cape Town.