Multiplier Effects of Energy Consumption and CO 2 Emissions by Input-Output Analysis in South Africa

This paper analyzed the energy consumption and CO 2 emission from 18 industrial sectors, and also evaluated the direct and indirect energy consumption and CO 2 emission of changes in the final demand of South Africa’s (SA) economy. To accomplish this goal, the input-output linkage and multiplier methods have been applied to investigate the interconnectedness of the 18 sectors’ input-output tables for the years 1995, 2000, 2005, 2010 and 2012, and to measure their total impact of energy commodity input coefficients and CO 2 emissions output coefficients for the year 2012. Results revealed that the electricity sector has a weak linkage with others sectors, which means it is mostly independent of other sectors. In another words, it does not induce and enable economic growth. Moreover, two sectors, such as Chemical and Petrochemical Industries and Basic Metals, were found as key sectors in SA’s economy in 1995, 2000 and 2012. In 2005 and 2010, only Chemical and Petrochemical Industries was the most important sector in SA. Additionally, Commercial and Public Services was the strongest forward linkage sector in SA. Our findings also showed that the electricity sector was the main direct monetary energy consumer and CO 2 emitter, and therefore the most dominant source in terms of energy and CO 2 intensities among all the 18 sectors in SA. Furthermore, our investigation of the direct and indirect effects on energy consumption and CO 2 emissions indicated that both total of direct energy consumption and CO 2 emissions were higher than both total indirect energy consumption and CO 2 emissions. Finally, some potential suggestions on reducing the energy consumption and CO 2 emissions deduced from this study are discussed.


INTRODUCTION
Energy plays a critical role in facilitating development in a country. Electricity is a form of energy which is essential in our daily life and is commercialized in other industry sectors. South Africa's industrial sector represented almost 30% of its gross domestic product (GDP, constant 2010) in 2014(World Bank, 2017. South Africa is the most developed and industrialized country in Africa. Its economy relies heavily on its energy sector, which is dominated by a huge supply of coal. Coal combustion is generally more carbonintensive than the burning of natural gas or petroleum for electricity. During the period 1990-2014, SA' s GDP, energy consumption and CO 2 emissions increased as shown on Fig. 1. The GDP rose to 412.10 billion USD from 223.04 billion USD, with an average annual growth of 2.61% (World Bank, 2017). The energy consumption and CO 2 emissions increased respectively from 2,137,277.66 Terajoules (TJ) to 3,130,721.57 TJ and 243.8 million tonnes of CO 2 (Mt CO 2 ) to 437.4 Mt CO 2 with a respective average annual growth of 1.66% and 2.55% (IEA data, 2016).
In 2014, SA's industrial sector had the highest share of energy consumption (37%) of the total energy consumption in the county. The transport, residential, commercial and public services, agriculture and non-specified sectors shared 24%, 23%, 6%, 3% and 2%, respectively (IEA data, 2017). Similarly, the industrial sector also led the consumption of electricity among all the other economic sectors, with 61% of the total, followed by the residential, commercial and public services, with 19% and 14%, respectively (IEA data, 2017). Furthermore, the electricity sector was the largest source of SA's CO 2 emissions, accounting for about 66% of SA's total, followed by the transportation sector with 13%, then, the industrial sector with 12%, next, the residential and commercial and public service sectors with 6%, and finally, the other sectors with 3%.
South Africa has an energy intensive economy with a high dependence on coal, as mentioned above. The country also  (2017)).
has one of the cheapest sources of energy, as this is viewed as a comparative advantage for economic development. South Africa committed to reduce its GHG emissions by 34% by 2020 and 42% by 2025 under the business-as-usual (BAU) levels. In 2015, South Africa (SA) also submitted its Intended Nationally Determined Contribution (INDC) report that included a goal for reducing its national emissions to 398 and 614 million tonnes of carbon dioxide equivalent in the years 2025and 2030(Department of Environmental Affairs, 2015. Therefore, South Africa is confronted with the impasse of simultaneously growing its economy and responding to international pressure to reduce GHG emissions. Policy makers need to promote options that benefit the environment without being harmful to economic growth and the path of development adopted. Different methods have been used to determine the strength of the causal link between energy consumption/CO 2 emission and economic growth (Climent and Pardo, 2007). For illustration, Muangthai et al. (2014) used Divisia index decomposition to evaluate the key influences affecting the evolution of CO 2 emissions from Thailand's thermal power sector during 2000-2011. The results indicated a coupling case between energy consumption and CO 2 emissions during 2000-2005, while a relative decoupling was observed for 2006-2011. Additionally, the economic effect was the main factor leading to increased CO 2 emissions from Thailand's thermal power generation, whereas electricity intensity played a major role in decreasing CO 2 emissions. Liou et al. (2015) modified conventional two-stage data envelopment analysis (DEA) to analyze the energy use efficiency and the economic efficiency of 28 Organization for Economic Co-operation and Development (OECD) countries from 2005 to 2007. Results indicated that OECD countries are only interested in economic development, have little concern for energy use efficiency, and continue to release considerable quantities of CO 2 . Lin et al. (2015) evaluated the decoupling of CO 2 emissions from GDP in South Africa using OECD and Tapio during the period 1990-2012, and they investigated the primary CO 2 emission drivers with the Kaya identity. Results showed that a strong decoupling occurred during 2010-2012, and the increase in population, GDP per capita, and adverse energy efficiency were the primary driving forces for increased CO 2 emission. Li et al. (2016) looked over challenges and perspectives of carbon fixation and utilization technologies. Results revealed that CO 2 fixation via fast-growing biomass decreases CO 2 uses for supplying chemicals and energy products, while integrated alkaline waste treatment with CO 2 capture and utilization is a smart approach to accomplishing direct and indirect reduction of GHG discharges from industries. Beidari et al. (2017) applied the Log Mean Divisia Index (LMDI) to investigate the influence of the factors which governed electricity generation-related CO 2 emission in SA over the period 1990-2013. The results showed that the electricity generation intensity effect was the most important contributor in decreasing CO 2 emissions. However, the effect of economic activity was the major factor that contributed to increasing CO 2 emissions.
The reasons for climate change are many and various.

GDP
Energy consumption CO2 emissions from fuel combustion Among them, these are the combustion of fossil fuels and also emission from the linkages between economic and industrial sectors. The objective of this paper is to analyze the energy consumption and energy emission from a number of economic industries, and also to evaluate the direct and indirect energy consumption and CO 2 emission of changes in the final demand in an economy. The application of Input-Output Analysis (IOA) is one of the tools widely used in the world. Many studies have applied IOA to environmental concerns related to the energy sector. Leontief (1970) was the first to introduce input-output analysis for calculating pollutant emission and assessing control approaches for major industries in the U.S.A. Lin et al. (2012) focused on the modeling of economic-based linkage effects of CO 2 emissions and the CO 2 multipliers from the electricity industry in Taiwan  The results indicated that the electricity generation sector has a high forward linkage effect and a fairly low backward linkage effect. The results also revealed that in 2010 the electricity generation sector was the highest energy intensive and CO 2 intensive industry.
Most of the previous studies in SA only focused on the linkage effects between economic sectors. They barely investigated the environmental impacts related to the interdependency of those sectors. For example, Stilwell et al. (2000) applied input-output methods to investigate the impacts of gold, coal, and other mining activities upon the South African economy during the period 1971-1993. The results pointed out that the linkages between mining and the rest of the economy were insufficient. Tregenna (2008) focused on the manufacturing sector by applying inputoutput tables to analyze inter-sectoral linkages in the South African economy. Her results demonstrated that the strong backward linkages from manufacturing to services showed that cost and quality of services inputs are critical for the competitiveness of manufacturing. Botha (2013) used IOA method to quantify and explain the total impact of South African Economy Industries in order to understand the interdependence of industries and how changes in one industry might affect other industries. Zhao et al. (2015) applied the environmental input-output model with the modified hypothetical extraction method to analyze the carbon linkage among sectors. The results indicated that the total carbon linkage of industrial systems in South Africa in 2005 was 171.32 million tonnes (Mt), which accounts for 81.58 Mt total backward carbon linkage and 89.71 Mt total forward carbon linkage. The results also revealed that the largest total carbon linkage and internal and net forward effect was attributed to the electricity sector.
The purpose of this study is to examine the interconnectedness of the electricity generation sectors for 18 sectors' input-output tables for the years 1995, 2000, 2005, 2010 and 2012, and to measure their total impact of energy commodity input coefficients and CO 2 emissions output coefficients for the year 2012. The electricity sector is the most prevailing source in terms of energy consumption and CO 2 emissions in SA, and also, one of the main sectors that influences its economic development. Since IOA is an effective approach that continues to grow in popularity as a method for assessing the relationship between economic activities of industrial sectors and embodied environmental impacts, this paper adopts the economic-based linkage effects of energy consumption and CO 2 emission of electricity production in SA. First, the linkage effects of the 18 aggregated sectors will be evaluated; then, their energy and CO 2 multipliers will be quantified; and finally the total ratio of direct and indirect effects energy consumption and CO 2 emissions will be carried out for the base year of 2005.

Input-Output Analysis
As mentioned by Lin et al. (2012) Leontief was the first to present the input-output model and its application to significant economic problems. The industries, households, and government entities are represented in the IO model in which the output of an industry can be considered as the input of other industries. Furthermore, due to the interdependence of sectors, the balances between the total input and the aggregate output of each product and service in a given economic system can be expressed by linear equations (Leontief, 1970). Based on Lin and Chang (1997), the basic equations of the IO can be expressed as following: where X i is the total gross output produced in sector i, X j is the total gross input required in sector j, F i is the product of sector i delivered to the final demand, V j is the final payment (value added) by sector j, x ij is the amount of the product sector i used by per unit of output of sector j a ij = x ij /X j is the direct input or technical coefficients from product sector i used by per unit of output sector j.
For example, input of aluminum (i) bought by car producers (j) in 2012 and total car production 2012 will give us the ratio of aluminum input to car output, x ij /X j [the units are ($/$)], and indicated it by a ij : value of aluminum bought by car producers in 2012 value of car production in 2012 This ratio is called a technical coefficient. So, if x 12 = $500 and X 2 = $20, 000 (sector 2 used $500 of goods from sector 1 in producing $20, 000 of sector 2 output), a 12 = x 12 / X 2 = $500 / $20,000 = 0.25. Since a 12 is actually $0.25/$1, the 0.25 is represented as the "dollars' worth of inputs from sector 1 per dollars' worth of output of sector 2." Namely that the technical coefficient represents each unit of the output value for the sector j that involved some amount of money to buy directly from sector i. Consequently, the technical structure of the entire system can be represented by the matrix of technical IO coefficients of all its sectors. In order to obtain the Leontief inverse matrix, Eq. (3) can be modified in the following matrix (Lin and Chang, 1997;Lin et al., 2012): where A is the direct input coefficient matrix of a ij , I is the identity matrix, B is the Leontief inverse matrix, and b ij are the elements of the Leontief inverse matrix which represent the total direct and indirect requirement of sector i by per unit of output sector j to final demand.

Linkage Effect Analysis
The idea of linkage effect was developed by Hirschman (1958). It is based on the assumption that the economy could be promoted by adopting an imbalanced investment policy to generate an equilibrium growth among the related industries. In other words, economy in related industries can be boosted through linking input/output activities.
In general, linkages have been categorized into two types: "backward linkage effect" and "forward linkage effect". Backward linkage effect refers to the demand-side connections an industrial sector has with other existing industrial sectors. From an input-output point of view, this linkage is the intermediate demand that an industrial sector makes on other industrial sectors. While forward linkage refers to the supply-side links an industrial sector has with other existing industrial sectors. For instance, many industrial sectors produce for inter-industry demand instead of final consumption. The calculation of the inter-industry linkage effect can be presented as following (Lin and Chang, 1997): According to Miller and Blair (2009), from the normalized form of the linkages, sectors can be classified in four ways as following: (1) mostly independent of (not strongly connected to) other sectors (both linkage measures less than 1), (2) mostly dependent on (linked to) other sectors (both linkage measures greater than 1), (3) dependent on inter-industry supply (only backward linkage greater than 1) and, (4) dependent on inter-industry demand (only forward linkage greater than 1).

Multiplier Analysis
The concept of multiplier was first applied by Wright (1974) to describe the energy commodity in input-output analysis. Lin and Chang (1997) stated that various types of pollutants can be associated with a measurable way to energy consumption or processes in the sense that pollution is assimilated to the "externalities" of steady economic activities. In this paper, the "externalities" are integrated into the formal input-output analysis. Also, Miller and Blair (1985) designed the energy and environmental input-output analysis to quantify the total impact of energy commodity input coefficients and pollutant output coefficients. In this study, the multipliers were calculated by the following equations: where M = [M j ] 1×n is the Energy Multiplier, total impact of energy coefficient, which specifies the amount of energy consumption required directly and indirectly caused by per $10 6 worth of output of industry j, (TJ/Million USD), m = [m j ] 1×n , is the energy consumption coefficient from industry j, (TJ / Million USD), , is the CO 2 Multiplier, the total impact of CO 2 emissions, which specifies the amount of CO 2 emitted directly and indirectly caused by per $10 6 worth of output of industry j, (kt CO 2 /Million USD), and q = [q j ] 1×n , CO 2 emissions coefficient from industry j, (kt CO 2 /Million USD).

DATA CONSOLIDATION
The purpose of this study is to examine the interconnectedness of the electricity generation sectors for 18 sectors of SA based upon input-output tables for the years 1995, 2000, 2005, 2010 and 2012, and to measure their total impact of energy commodity input coefficients and CO 2 emissions output coefficients for the year 2012.
Input-output data for 1995, 2000, 2005 and 2010 (which include 34 sectors) was obtained from the Organization for Economic Cooperation and Development (OECD) inputoutput database (with 2011 being the latest one). In that regard, the 2012 I-O database, with 50 sectors was downloaded from Statistics South Africa (SSA, 2017) website database. Statistics South Africa database provides input-output tables from 2009 to 2012 (the most recent one). The energy consumption in (TJ) and CO 2 emissions from fuel combustion in (kt CO 2 ) data of Fig. 1 were collected from the International Energy Agency (IEA, 2017) statistics website. The GDP (constant 2010) was measured in USD, and the total population was from the World Bank (2017) website. The statistical energy consumption data in (TJ) for years 1995,2000,2005,2010 and 2012 was collected from Energy Balance Sheets of SA Department of Energy (DOE, 2017) statistics website; the CO 2 emissions in (kt) from energy consumption of each sector were calculated following the revised guidelines 2006 IPPC (IPCC, 2006). Finally, as the sector classifications of the IO table from OECD, Statistics South Africa and those from the Energy Balance Sheets of DOE are not homogenous, we combined all the IO economic tables and energy balance tables into 18 sectors to ease the calculation and interpretation of results, as defined in Table 1.

Inter-Industry Linkages
Tables 2 (1995,2000,2005,2010,2012); ** Economic Cooperation and Development (1995,2000,2005,2010); *** Statistics South Africa (2012). Residential sector in 1995 and 2010. The remaining sectors were those with weak linkages (both linkage measures less than 1) which include the Electricity sector. An increase in the final demand of the key sectors' output will have a huge impact on sectors that supply inputs in the production of these key sectors' output. In other words, the key sectors can induce and support other sectors in economic development. The strong backward sectors generate supplementary demand for the output of upstream sectors, leading to increased upstream output, capacity utilization and upstream technological advancement. On the other hand, the strong forward sectors provide the needs for downstream sectors, comprising downstream investment or technological advancement. In other words, the strong backward sectors influence economic development while the strong forward sectors support economic development (Hirschman, 1959). The weak linkage sectors are mostly independent of other sectors; that means they do not induce nor enable economic growth. The top 5 backward and forward linkage sectors are presented in Tables 4 and 5. Commercial and Public Services was the sector with the highest forward linkage for each year of this study. It has proved to be the major support of the country's economic growth (over 60% of SA's GDP) over the years (World Bank, 2017). This sector boasts other sectors by providing a full range of services such as commercial, retail and merchant banking, mortgage lending, insurance, investment, etc…. It is noticeable that electricity sector has a weak linkage and neither influences nor supports the economic development in South Africa according to the IOA results. In fact, as mentioned by Beidari et al. (2017), the electricity sector in SA is facing many challenges due to the increase of the price of electricity, which can be related to the aging of most power stations as they approach the end of their lifespan, causing substantial operational inefficiencies. Furthermore, since the initial electricity blackouts in 2008, Eskom (SA's national electricity distributer) is struggling to meet the country's electricity demand. That will inevitably induce a decrease in total output of the industrial sector (largest electricity consuming sector) followed by the residential, and commercial and public service (second and third largest electricity consuming sectors). The inter-industry linkages of the Electricity sector were illustrated in Table 6

Energy Multiplier
The results of energy consumption, monetary energy consumption and energy multiplier of the 18 industrial sectors in 2012 are shown in Table 7. Electricity production, Gas and Water Supply, Other industries, Transport, Basic Metals and Residential are the top 5 sectors in terms of energy consumption and monetary energy consumption in 2012 in SA. Regarding the energy multiplier in 2012, the top 5 sectors are Electricity, Gas and Water Supply; Other Industries; Transport; Basic Metals; and Chemical and Petrochemical Industries. Electricity, Gas and Water Supply is the sector with the highest value of energy consumption (56% of the total), monetary energy consumption and energy multiplier in 2012. However, its gross output is very low (3% of the total). That can explain why its energy intensity and energy multiplier are the highest among other sectors. Apart from the energy multiplier values of Electricity, Gas and Water Supply; Other Industries; Transport; and Residential, the results reveal that each value of all other sectors energy multiplier is more than double their monetary energy consumption respectively. It indicates that those sectors have large indirect energy consumption. This paper also compares the sectors' gross output and their energy intensity related rankings. The highest gross output sectors such as Commercial and Public Services, Non-specified (other sectors), Mining and Quarrying, Chemical and Petrochemical Industries, Food and Tobacco and Construction with a lower energy intensity are those industries which deserve a continuous growth of the economy. On the other hand, those with less economy such as Electricity, Gas and Water Supply, Residential and Other Industries, which demand more energy and are ranked the highest in energy intensity deserve attention for industries structure and/or the fuel structure shifting.

CO 2 Multiplier
From Table 8, we can see the results of CO 2 emissions, monetary CO 2 emissions factor and CO 2 multiplier in 2012 of the 18 industrial sectors in 2012. Similarly, Electricity, Gas and Water Supply; Other Industries; Transport; Basic Metals; and Residential sectors are the highest CO 2 emissions and monetary CO 2 emissions factors. This is quite logical insofar as if a sector consumes a huge amount of fossil fuel energy, it will probably emit a huge amount of CO 2 as well. The major sectors with the highest values of CO 2 multiplier in 2012 are Electricity, Gas and Water Supply; Other industries; Transport; Basic Metals; and Non-Metallic Products. The Electricity, Gas and Water Supply sector still remains is the dominant direct CO 2 emitter. The reason being that more than 90% of the energy consumed to generate electricity comes from coal. As in energy multiplier section, except for Electricity, Gas and Water Supply; Other Industries; Transport; and Residential, the results reveal that other sectors' CO 2 emission multipliers are more than double their monetary CO 2 emissions. This indicates that the indirect CO 2 emissions of the related sectors are larger than the direct ones. Similarly, the highest gross output sectors such as Commercial and Public Services, Nonspecified (other sectors), Mining and Quarrying, Chemical and Petrochemical Industries, Food and Tobacco and Construction with a lower CO 2 emissions intensity are those industries which deserve a continued growth of the economy. On the other hand, sectors such as Electricity, Gas and Water Supply; Residential; and Other industries, which are ranked the highest in CO 2 emissions intensity deserve attention for industries structure and/or the fuel structure shifting. R* = Sectors 'gross output ranking; R** = Sectors' energy intensity ranking.

Direct and Indirect Effects of Energy Consumption
The calculated results of the direct and indirect effects of energy consumption of the 18 sectors are presented in Table 9. The majority of sectors have a higher indirect effect of energy consumption except for four sectors which are Electricity, Gas and Water Supply; Transport; Other Industries; and Residential. This shows that most of their energy consumption comes from the Energy sector. Electricity, Gas and Water Supply has the highest direct effect on energy consumption (83.84%) and the lowest indirect effect (16.16%). This clearly indicates that the electricity generation sector relies less on other sectors for its fuel inputs for electricity generation. Finally, the total direct effect of energy consumption of the 18 aggregated sectors accounts for 50.15%, while 49.85% is attributed to the total indirect effect of energy consumption. This can be explained by the fact that the majority of the energy consumption is directly consumed by the electricity sector. In fact, according to the result in Table 7, electricity sector totals of energy consumption and monetary energy consumption (direct energy consumption) account respectively 56% and 55% of the totals of energy consumption and monetary energy consumption, respectively.

Direct and Indirect effects of CO 2 Emissions
Table 10 presents the results related to the direct and indirect effects of CO 2 emissions of the 18 sectors in SA for the year 2012. It is quite similar to the results of direct and indirect energy consumption. As we know, the volume of CO 2 released from fossil fuel combustion depends principally on the type and quantity of fuels used. Therefore, the sectors with direct effects of energy consumption that are higher than their indirect effects of energy consumption also have greater direct effects of CO 2 emissions and vice versa. Electricity, Gas and Water Supply; Transport; Other Industries; and Residential are the sectors with greater direct effects of CO 2 emissions, and the other sectors come with higher indirect effects of CO 2 emissions. Moreover, the total direct effect of CO 2 emissions intensity of the aggregated 18 sectors accounts for 51.05%, while the total indirect effect of CO 2 emissions intensity accounts for 48.95%, which is quite reasonable. The main reason might be that among all the 18 sectors, the electricity sector has the highest direct effect (84.06%) and the lowest indirect effect (15.94%), as presented in Table 10; and also, according to the result in Table 8, electricity sector totals of CO 2 emissions and monetary CO2 emissions (direct CO 2 emissions) account respectively 58% and 55% of the totals of CO 2 emissions and monetary CO 2 emissions, respectively. This is well correlated with the introduction section and the previous study of Beidari et al. (2017) where it is stated that the electricity sector was the largest source of SA's CO 2 emissions, accounting for about 66% of the SA total and more than 90% of the electricity generated in South Africa comes from coal. Those sectors with a large indirect effects need more attention. If we only count the direct emissions of those sectors with a large indirect CO 2 emission, the total CO 2 emissions will be underestimated. If also the total indirect CO 2 emissions was higher than the total direct CO 2 emissions, the total CO 2 emissions could be underestimated. Nonetheless, some enhancements can be made through policy adjustment to reduce high amounts of indirect energy consumption for related sectors, which will reduce their indirect CO 2 emissions.

Policy Implications
Based on the results of this study, the following propositions can be useful to enhance the electricity sector's linkage effects in order to become a key sector for SA, and reduce its direct energy consumption and CO 2 emissions, SA's government should: (1) invest more in the electricity sector in order to expand electricity capacity. That means they will need to build some new electric power stations.
(2) encourage public investment in order to maintain and expand electricity capacity. That would help to relieve Eskom (the national electricity distributer, which produces 95% of SA's electricity), because Eskom's plant is under severe stretch due to factors such as poor quality of coal, staff deficiencies and a high load on its capacity (Bayliss, 2008). (3) shift the fuel structure especially in the electricity sector by developing more the renewable energy sector. (4) increase the quality of coal and provide new technologies such as Carbon Capture and Storage as mentioned by Beidari et al. (2017) for the main power plants. That could inevitably increase the efficiency of power plants.
That means less coal could be used to produce the same amount of energy or more.
(5) reduce the amount of its oil imports by expanding its renewable energy capacity.

CONCLUSIONS
This paper has applied input-output analysis to investigate the interconnectedness of 18 aggregated sectors for the years 1995, 2000, 2005, 2010 and 2012, and adopted multiplier analysis to quantify the total environmental impacts related to the inter-industry linkages for the year 2012 in South Africa, with a focus on the relationship between the electricity sector and the rest of the economy.
First, the linkage effects were calculated using the PIOD, which stands for backward linkage, and SIOD, which stands for the forward linkage. Results point out that the electricity sector has a weak linkage (both backward and forward linkages are less than 1) with others sectors, which means it is mostly independent of other sectors. In another words, it does not induce and enable economic growth by IOA. Moreover, two sectors such as Chemical and Petrochemical Industries and Basic Metals were found as key sectors in SA's economy in 1995, 2000. In 2005and 2010, only Chemical and Petrochemical Industries was the most important sector in SA. Additionally, Commercial and Public Services was the strongest forward linkage sector in SA, which means it creates a large impact on downstream sectors, comprising downstream investment or technological advancement. In other words, it is the sector which supports the most economic development in SA.
The multiplier analysis was applied to map out the 18 sectors' energy and CO 2 emissions intensities for the year 2012. The results clearly showed that Electricity, Gas and Water Supply; Other Industries; Transport; Basic Metals; and Residential were the top five energy consuming sectors and CO 2 emitters in 2012 in SA. However, the electricity sector was the main direct monetary energy consumer and CO 2 emitter, and therefore it is the most dominant source in terms of energy and CO 2 intensities among all the 18 sectors in SA.
Furthermore, the results of multipliers analysis indicated that most total of indirect energy consumption and CO 2 emissions were higher than direct energy consumption and CO 2 emissions. This means that both indirect energy consumption and CO 2 emissions has large impacts on SA's energy consumption and CO 2 emissions, which cannot be ignored. Therefore, the SA government should plan to implement practical strategies to reduce the indirect energy consumption intensity as well as the indirect CO 2 emissions intensity.
Based on the results of this paper, a variety of propositions which can be useful to improve the electricity sector's linkage effects in order to become a key sector for SA, and reduce its direct energy consumption and CO 2 emissions, have been recommended in the section of policy implications as suggestions.
Finally, this study showed that input-output analysis can be a very useful tool for governments, as it provides valuable information about the functioning and structure of the linkages among sectors in the national level economy, as well as access to both direct and indirect effects related to energy consumption and CO 2 emissions. The multipliers analysis applied in this study gives a good understanding of the interconnectedness among industries for a government to further evaluate the profiles of the direct and indirect effects among industries regarding their energy consumption and CO 2 emissions. Therefore, it can be a worthwhile tool to assist policymakers in elaborating appropriate economic policy as well as improving energy policy makings.