There exists a huge international literature on the, so-called, Environmental Kuznets Curve (EKC) hypothesis, which in turn, postulates an inverted u-shaped relationship between environmental pollutants and output. The empirical literature on EKC has mainly used test for cointegration, based on polynomial relationships between pollution and income. Motivated by the fact that, measured in per capita CO2 equivalent emissions, South Africa is the world's most carbon-intensive non-oil-producing developing country, this paper aims to test the validity of the EKC for South Africa. For this purpose, we use a century of data (1911-2010), to capture the process of development better compared to short sample-based research; and the concept of co-summability, which is designed to analyze non-linear long-run relations among persistent processes. Our results, however, provide no support of the EKC for South Africa, both for the full-sample and sub-samples (determined by tests of structural breaks), implying that to reduce emissions without sacrificing growth, policies should be aimed at promoting energy efficiency. (C) 2015 Elsevier B.V. All rights reserved.
