by Monkgogi Bonolo Otlhogile
With an average growth rate of 4.3% between 2001 and 2007, South Africa joined Brazil, Russia, India and China as the fifth member of BRICS, an association for the five emerging economies of the world in 2010. However, South Africa also joined these countries as one of the major carbon dioxide emitters, producing 1% of the world’s emissions. The environmental Kuznets curve (EKC) hypothesis states that early economic development will result in an increase in environmental degradation. This includes pollutants such as carbon dioxide and sulfur, which are considered by-products of economic activity. Eighty-seven percent of the carbon dioxide emitted by South Africa is a by-product of coal-fueled energy production used for industrial purposes, suggesting that South Africa’s economic development follows the EKC hypothesis. However, the hypothesis also states that as an economy matures, environmental degradation is reduced by investments in cleaner production technologies and practices. Shahbaz et al. (2013) argue that financial development could mitigate environmental degradation because mature financial institutions encourage investments in environmentally friendly projects. To test this theory and the environmental Kuznets curve hypothesis, the authors assessed the effects of economic growth, coal consumption, trade openness, and financial development on carbon dioxide emissions in South Africa over the period 1965–2008. After testing for dynamic changes in the data, Shahbaz and colleagues used an autoregressive distributed lag (ARDL) bounds testing approach to test the relationship among the above variables in the long run, and used the error correction method (ECM) to test the relationship in the short run. Shahbaz et al. found that coal consumption and economic growth increased environmental degradation, verifying the existence of an EKC curve in South Africa. However, the authors found that trade openness and financial development decreased energy emissions and pollutants. This suggests that South Africa could mitigate economic-related carbon dioxide emissions and sustain economic growth by reforming trade policy and their financial systems as they could facilitate efficient energy use.
To test the effects of financial development, economic growth, coal consumption, and trade openness on carbon dioxide emissions and to validate the existence of the environmental Kuznets curve, Shahbaz and colleagues in India and Pakistan developed a model composed of multiple tests: the Saikkonen and Lütkepohl structural break unit root test, an autoregressive distributed lag (ARDL) bounds testing approach, a Granger causality analysis, an error correction method (ECM), and diagnostic tests to verify the stability of their model. In their model, the authors used carbon emissions per person or per capita as an indicator of environmental degradation and gross domestic product (GDP) per capita to measure economic growth. Financial development was proxied by access to domestic credit per capita in the private sector while trade openness was quantified by dividing exports and imports by GDP. They also added coal consumption per capita and urbanization rates to their model. The authors used data from the World Bank Indicators for the above stated variables and established the directionality of their variables using economic and environmental theories to produce a set of expected possible outcomes. They expected an accumulation of energy pollutants and a rise in economic growth to result in increased carbon dioxide emissions as stated in the EKC hypothesis. The authors also expected a mature financial sector to decrease carbon dioxide emissions via environmentally friendly resource reallocation and by incentivizing the use of cleaner technologies and practices. While an immature financial sector would exacerbate carbon dioxide emissions as it would be more interested in profit. Therefore the model would actually determine the maturity of the financial sector in South Africa. Because South Africa relies on coal for 77% of its energy needs, Shahbaz et al. expected increasing coal consumption to result in increasing carbon dioxide emissions. The effect of trade openness on carbon dioxide emissions was ambiguous because of scale, technique and composition effects. The scale effect suggests that the level of trade liberalization exacerbates emissions due to economic growth while the technique effect suggests that importing more efficient technologies reduces emissions. The composite effect suggests that trade liberalization may increase or decrease emissions depending on the industries the country has a comparative advantage in. The strength of the effects on South Africa’s trade openness would determine the increase or decrease of carbon emissions. The authors tested their expected outcomes over the period 1965–2008.
When an analysis is done on data that has been collected over time (known as a time series analysis), the relationship among variables may change because of dynamic events such as the end apartheid that cannot be modeled. However, there are tests such as the Ng and Perron test that allow economists to detect these structural breaks within their data. The Ng and Perron test showed that there were structural break points in the data, but an Ng and Perron test will not find the dates where variable relationships change or shift. The authors therefore used the Saikkonen and Lütkepohl structural break unit root test to isolate the specific dates of structural break points. The test showed intermittent changes in the relationship between carbon dioxide emissions, economic growth, financial development, coal consumption, and trade openness from 1965–2008 especially in the years 2000 and 2004. By pin-pointing the years of structural breaks, the authors were able to perform an autoregressive distributed lag (ARDL) bounds testing approach to cointegration which is a more robust test of relationships between variables as it allows for a error correction model which tests short-run dynamics within a long-run equilibrium. The ARDL approach tests the long and short-term relationship between a variable such as carbon emissions and economic growth by predicting the values of one variable in the current period by using current and past values of another variable. The results from the Saikkonen and Lütkepohl structural break unit root tests therefore allowed the authors to pick specific time periods that did not interfere with the structural breaks in the data. The authors were careful to state that the results from the ADRL cointegration tests depicted a correlation between any two variables but it did not necessarily mean that one variable caused an increase or decrease in another. To test the direction of the causal relationships between the variables, Shahbaz et al. used a Granger causality test. To test the stability of their model in the long and short run, they ran a set of diagnostic tests such as the ARCH and normality tests. The model in the short and long run passed all its diagnostic tests proving it to be statistically stable.
The ARDL testing approach found that there is a long run relationship between economic growth, financial development, coal consumption, trade openness, and carbon dioxide in South Africa. The results showed that in the long run, carbon emissions in a past period contribute to carbon dioxide in the next period, which means there is inertia in emissions. In addition, the authors found that in the short run and long run a 1% increase in GDP is linked to a 0.2566% and 0.223% increase of carbon dioxide emissions, respectively. This suggests that South Africa’s economic growth has come at the expense of environmental quality. Interestingly, the authors also found that economic growth had both positive and negative effects on carbon dioxide emissions over time which validates the EKC hypothesis as it proves that South Africa’s early economic development caused environmental degradation while its stabilization at a threshold of $3463 GDP per capita caused the decline of environmental degradation. The Granger causality test proved that economic growth directly caused changes in carbon dioxide emissions, which also confirmed the existence of an EKC in South Africa. Shahbaz and colleagues believe that environment degradation was caused by the positive association between coal consumption and carbon dioxide emission because of South Africa’s heavy reliance on its coal reserves but could also be exacerbated by urbanization, which also increases energy consumption. This hypothesis was validated by the ARDL test both in the short and long run while the Granger causality test proved that increased coal consumption was causing increased carbon dioxide emissions. The ARDL results showed that a 1% increase in coal use increased carbon dioxide emissions by 0.569% in the long run, which was further confirmed by the results of the short run model. As the authors had posited, financial development had a negative effect on carbon dioxide emissions in both the long and short run. Results in the long run show that a 0.00273% reduction in carbon dioxide emissions would stem from a 1% in financial development. The negative effect proves that South Africa’s financial systems are mature enough to allocate resources efficiently and to encourage clean technology, thus improving environmental quality. Trade openness also had a negative effect on carbon dioxide emissions in both the long and short run, though the effect was not statistically significant in the short run. This suggests that trade openness improved environmental quality, however, the authors were not able to isolate whether this was a result of technical, composite, and/or scale effects as discussed above.
The greatest sources of environmental degradation in the 21st century are greenhouse gases (GHGs), however, carbon dioxide emissions have become a salient issue because of the global effects these emissions have, irrespective of the emitter. As one of the emerging economies and major emitters of carbon dioxide, Shahbaz and colleagues’ study proves that South Africa’s economic development can be modeled by the environmental Kuznets curve. However, the alleviating effects of trade openness and financial development on carbon dioxide emissions suggest that there are policy adjustments that could be made to decrease emissions in South Africa. The authors’ study provides empirical proof that financial and trade reforms could improve environmental quality by increasing foreign and domestic investments in environmentally friendly project, increasing energy efficiencies, and the development of new technologies which would result in a less carbon-intensive South African economy which would be better for everyone.
Shahbaz, M., Tiwari, A., Nasir, M., 2013. The Effects of Financial Development, Economic Growth, Coal Consumption and Trade Openness on Environmental Performance in South Africa. Energy Policy 6, 1452–1459.