TESTING FOR LONG-RUN ELASTICITIES IN THE PRESENCE OF STRUCTURAL CHANGE: EVIDENCE FROM CLIMATE CHANGE FOR SOUTH AFRICA

Abstract

This study uses Dynamic Ordinary Least Squares (DOLS) developed by Stock and Watson (1993), Fully Modified Ordinary Least Squares (FMOLS) developed by Phillips and Hansen (1990), and Canonical Cointegrating Regression (CCR) developed by Park (1992) methods to provide insights into the responsiveness of climate change to changes in energy consumption, urbanization, and economic growth over a significant period. The DOLS, FMOLS, and CCR methods serve as supplementary robustness checks to the existence of a long-term relationship ascertained among the stated variables under study (Auwal, 2024). The study detects a structural change in the long run in the effect on the South African economy from 1971 to 2022, using the least squares method and Bai & Perron's (2003) tests, which necessitates the use of a dichotomous variable to represent the categorical data periods of investigation. The findings from the DOLS, FMOLS, and CCR models revealed are three-fold. On energy consumption, climate change significantly and positively responds to changes in coal and oil consumption by 0.85% to 1.1% and 0.25% to 0.27%, respectively. While the degree of responsiveness of climate change to gas consumption is significantly negative between 0.04% and 0.15%, the climate change response to changes in urbanization and economic growth was insignificant. The study concludes that while urbanization can create localized climate changes, it is not considered to be a significant direct contributor to climate change in South Africa on a large scale. Similarly, its economic activity alone doesn’t significantly influence the progression of climate change in South Africa.