تولید گازهای گلخانه ای CO2، مصرف انرژی، درآمد و تجارت خارجی: چشم انداز آفریقای جنوبی

The effect of trade liberalisation on environmental conditions has yielded significant debate in the energy economics literature. Although research on the relationship between energy consumption, emissions and economic growth is not new in South Africa, no study specifically addresses the role that South Africa's foreign trade plays in this context. A surprising fact given trade is one of the most important factors that can explain the environmental Kuznets curve. This study employs recent South African trade and energy data and modern econometric techniques to investigate this. The main finding of interest in this paper is the existence of a long run relationship between environmental quality, levels of per capita energy use and foreign trade in South Africa. As anticipated per capita energy use has a significant long run effect in raising the country's CO2 emission levels, yet surprisingly higher levels of trade for the country act to reduce these emissions. Granger causality tests confirm the existence of a positive bidirectional relationship between per capita energy use and CO2 emissions. Whilst the study also finds positive bidirectional causality between trade and income per capita and between trade and per capita energy use, it appears however that trade liberalisation in South Africa has not contributed to a long run growth in pollution-intensive activities nor higher emission levels.

If South Africa's greenhouse gas (GHG) emissions are compared on a global scale, it is immediately clear that the country is one of the world's most carbon-intensive economies. In fact, South Africa is the world's most carbon-intensive non-oil-producing developing country, measured in per capita CO2 equivalent emissions in 2010, and excluding island states (Energy Information Administration (EIA), 2010). Furthermore, it is the largest emitter of GHGs in Africa, with 42% of the continents emissions coming from South Africa alone. South Africa is also a bigger emitter of CO2 than all other Sub-Saharan African (SSA) countries combined (Energy Information Administration (EIA), 2010). South Africa's total GHG emissions in 2000 were estimated to be 461 million tons CO2 equivalent of which, 83% of emissions were associated with energy supply and consumption, 7% from industrial processes, 8% from agriculture, and 2% from waste (Department of Environmental Affairs (DEA), 2010). The energy sector is therefore by far the largest sector responsible for emissions in South Africa at 380,988 Gg CO2e with the sector's combustion of fuel producing 81% of the sector's emissions and fugitive emissions from fuel contributing the remaining 19% (Department of Environmental Affairs (DEA), 2010). Factors which have contributed to South Africa's enormous energy related emissions include: a deliberate strategy by the pre-democratic government prior to 1994 of encouraging investment in energy-intensive industries, including aluminium and other non-ferrous metal beneficiation (the so called ‘mineral-energy complex’ identified by Fine and Rustomjee (1996)); and the carbon-intensity of a largely (90% +) coal-based electricity generation base (Energy Information Administration (EIA), 2010). Of particular relevance to our work is a finding in an multi-country study on CO2 embodied in international trade by Peters and Hertwich (2007) that around 40% of South Africa's emissions are due to trade (in particular the export of carbon-intensive goods) rather than domestic consumption. According to this study, this is the highest proportion for any of the countries included in their analysis. Given that most of South Africa's energy needs are met by burning fossil fuels, a strong link between foreign trade and CO2 emissions is to be expected. This fact makes South Africa vulnerable to trade-induced environmental degradation as a result of an increase in the burning of fossil fuels to meet the energy demands of an expanding export sector. This is particularly relevant given international trade is one of the most important factors that can explain the environmental Kuznets curve (EKC). In this context, it is important to note that South Africa is both a member of the World Trade Organisation (WTO) since 1995 and a signatory to the 1992 UNFCCC and its Kyoto Protocol. Under Kyoto, the biggest emitters of GHGs are encouraged to implement mitigation measures that catalyse energy efficiency and motivate energy sustainability policies. South Africa is classified as a non-annex developing country and therefore has no mandatory emission reduction targets during the period 2008–2012. Nonetheless, the country is committed to the fight against climate change and has instituted several policies and strategies at the national level to reduce GHG emissions. The South African government is of the view that the country needs to reduce GHG emissions while working to ensure economic growth, increase employment, and reduce poverty and inequality (National Treasury, 2010). The above discussion suggests South Africa is a compelling candidate for a separate study which investigates the role of trade openness on economic growth, energy consumption and pollutant emissions. Indeed, our study is the first attempt to incorporate foreign trade as a separate determinant of CO2 emissions in a multivariate framework in the context of South Africa. Whilst there are some studies in the international literature that link economic growth, energy consumption and pollutant emissions in the same framework initiated by the work of Ang (2007) and Soytas et al. (2007) these do not relate to South Africa. Studies that do focus on South Africa investigate either the link between economic growth and emissions (see, Nahman and Antrobus, 2005); or energy consumption and economic growth (see, Ziramba, 2009, Odhiambo, 2009, Wolde-Rufael, 2006 and Wolde-Rufael, 2009). The only South African study that our research has revealed that employs modern advances in time series econometrics of co-integration and causality to test the relationship between energy consumption, pollutant emissions and economic growth in a coherent multivariate framework is that of Menyah and Wolde-Rufael (2010). This paper's aim is to fill a gap in research in the South African context by employing the same econometric techniques of Halicioglu (2009) but in addition introducing foreign trade into the analysis as in the work by Baek et al. (2009). To our knowledge this is the first South African study that attempts to specifically understand the role of foreign trade on pollutant emissions through its effect on economic growth and energy consumption in a multivariate framework employing a single cointegration approach. The remainder of the paper is structured as follows. In Section 2 a brief review of the empirical literature is presented followed by a discussion in Section 3 of the data and methodology used. Section 4 presents the empirical evidence, while the concluding comments are outlined in Section 5.

The results of this study support the existence of a long run relationship between environmental quality, energy use and foreign trade in the case of South Africa. Specifically the study finds that South Africa's CO2 emission levels increase in the presence of greater per capita energy use within the economy yet decrease as the country's levels of foreign trade levels as a portion of national income rise. This finding must be viewed in the context that the effect of trade liberalisation on income, energy consumption and the environment is essentially an empirical phenomenon (Baek and Kim, 2011). Ultimately, as these authors point out, the outcome depends on the various characteristics of the economy under consideration such as: stage of economic development, openness levels and stringency of environmental regulations. Shahbaz et al. (2011) and Bildirici et al. (2012) confirm the study's finding that trade liberalisation in the context of South Africa is beneficial for the environment whereas Baek and Kim (2011) find the opposite and report that CO2 emissions have a positive long-run relationship with South African trade openness. By way of a bounds F-test based on the Wald's statistic the study confirms the existence of three long-run cointegrating relationships and that there are multiple variables influencing South Africa's CO2 emission levels. The absence of a statistically significant long run relationship between growth in income per capita and CO2 emissions in our model is most likely due to the endogeneity between foreign trade, income per capita and South Africa's emission levels. As such this result should not be interpreted to imply either the absence or presence of an EKC finding in the case of South Africa but rather that the precise relationship between the country's emissions and economic growth is difficult to identify in the presence of foreign trade. What is clear according to a very recent OECD study of South Africa is that as a result of the under-pricing of electricity and coal over the last two decades, the country's greenhouse gas emissions per unit of GDP are among the highest in the world and that South Africa has seen less decoupling of real GDP and CO2 emissions in recent years than most other countries ( OECD, 2013). Whilst our study also finds positive bidirectional causality between foreign trade and income per capita and between foreign trade and per capita energy use, it appears however that trade liberalisation in South Africa has not contributed to a long run growth in pollution-intensive activities nor higher emission levels. Rather, our research suggests that a higher degree of trade openness possibly reduces CO2 emissions in South Africa through an environment which stimulates technological innovations by increasing spending on energy R&D which results in energy efficiencies, fewer pollutants and hence environmental conservation. Indeed Shahbaz et al. (2011) argue that a higher degree of financial system development and trade openness prop up technological innovations in South Africa by (1) strengthening the institutional framework which creates incentives for firms to act in an environmental conscious manner and by (2) increasing spending on energy conservation R&D which results in energy efficiency and hence lowering emissions as suggested by Blanford (2009). In terms of energy R&D expenditures in South Africa, Pouris (2010) finds total expenditures have been on an increasing path in recent years. The study reports data published by South Africa's Department of Science and Technology on total energy R&D expenditures of R334 million in 2001/2 and R922 million in 2006/7. Pouris (2010) however argues that in comparison to the rest of the world the country under-spends in the energy field. In 2006, Japan had the highest intensity spending 3.32% of GDP on energy related research, whilst South African energy intensity was only 0.05% of GDP and the EU-27 had an average intensity of 1.86% of GDP (Pouris, 2010). More recently South Africa's media has quoted the senior general manager of the state-owned energy provider Eskom to have identified the lack of R&D spend to develop affordable water- and emissions-efficient technologies as the biggest concern for the country's energy consumers (Blaine, 2013). What is clear according to Menyah and Wolde-Rufael (2010) is that research and investment in clean energy should be an integral part of the process of controlling GHG emissions and finding alternative sources of energy to coal in South Africa. Improving energy efficiency and the exploitation of the country's renewable energy resources are identified in their study as viable options for meeting the significant energy and environmental challenges facing the country. In terms of the specific relationship between energy consumption and economic growth/openness, Bildirici et al. (2012) find evidence to support the energy conservation hypothesis in the case of South Africa. In particular, their study indicates that since economic growth induced by trade liberalisation causes an increase in South Africa's energy consumption, the policy of conserving energy consumption may be implemented with little or no adverse effect on economic growth and the further expansion of trade as is the case in a less-energy dependent economy than South Africa. The Bildirici et al. (2012) study supports the finding by Inglesi-Lotz and Pouris (2012) that South Africa's energy consumption has increased sharply since the early 1990s and that the main driver of this increase between the years 1993–2006 was due to output or production related factors, with structural changes playing a secondary role. Furthermore it is emphasised by Bildirici et al. (2012) that this effect is to be understood in the context that South Africa has undergone major political, social and economic changes after 1990, resulting in a sharp increase in economic activity. Inglesi-Lotz and Pouris (2012) confirm that structural changes in the South African economy have contributed to the increase in the country's energy consumption. Their study indicates that a decomposition of the country's energy efficiency is instructive, in that whilst South Africa's energy efficiency improves up to 2006, structural changes move it in the opposite direction. In the presence of the country's highly energy intensive economic activities and given South Africa's extreme dependence on coal as its major energy source, it is indeed surprising that the country's pollutant emissions outcome is not exaggerated in the presence of foreign trade. Indeed the preceeding discussion of the results of other studies in the South African context seem to suggest that this study's finding of a beneficial environmental outcome in the presence of increased South African foreign trade should be interpreted to imply that the impact of foreign trade on CO2 emissions is driven largely by scale and technique effects rather than structural composition of production effects for the country. This econometric finding should however be interpreted with care, as it may not be sufficiently robust enough to categorically state.