Impact of Economic Growth on Greenhouse Gas (GHG) Emissions—Social Accounting Matrix (SAM) Multiplier Analysis

Abstract

In general, direct pollution effect is only a small part of the total effect of pollution when production/consumption activities take place in an economy. To estimate the same, we have to undertake multiplier analysis. In this chapter, we have followed the social accounting matrix (SAM) multiplier method to estimate empirically the multiplier impact of economic activity on greenhouse gas (GHG) emissions in India. To be specific, this chapter explains the linkage of output growth with energy demand and GHG emissions. Finally, the issue of green employment opportunity has been analyzed with the help of SAM multiplier.

In general, direct pollution effect is only a small part of the total effect of pollution when production/consumption activities take place in an economy. To estimate the same, we have to undertake multiplier analysis. In this chapter, we have followed the social accounting matrix (SAM) multiplier method to estimate empirically the multiplier impact of economic activity on greenhouse gas (GHG) emissions in India. To be specific, this chapter explains the linkage of output growth with energy demand and GHG emissions. Finally, the issue of green employment opportunity has been analyzed with the help of SAM multiplier.

4.1 Approach to Link Economic Growth with GHG Emissions

Achieving economic growth, alleviating poverty by creating employment opportunity, and reducing GHG emissions are key policy issues for the late-industrializing economy like India. In this context, since energy demand is crucial for GHG emissions, the Government of India has taken various policy measures in its 12th 5-year plan (2012−2017) to improve energy efficiency in the industries (Planning Commission 2011). Again the SAM of the year 2006−2007 presented in Chap. 2 shows that the energy demand pattern is different for different sectors. Hence, the impact of growth in economic activities will have different impact on energy demand and hence GHG emissions.

Since then, a SAM multiplier describes growth in economic activity due to exogenous policy shock into the economy, linking SAM multiplier with the sector-specific energy and GHG emissions intensities; one can estimate the multiplier impact of economic growth on energy use and GHG emissions. Like a SAM multiplier, this multiplier impact will also include the direct, indirect, and induced impact of economic growth on GHG emissions while there is any exogenous change in the economy (Robert 1975).

In this study, we have assumed Government Account, Account of Net Indirect taxes, and Foreign Trade accounts of our SAM as exogenous. Keeping these accounts as exogenous, we have done the following impact analysis:

• 1. The impact on energy use due to growth in sectoral output resulting from any exogenous changes in the economy

• 2. The impact on GHG emissions due to growth in sectoral output resulting from any exogenous changes in the economy

Apart from these linkages between output growth, energy consumption, and GHG emissions, linking GHG emissions with employment generation is also crucial from the point of view of green growth. Recently, the concept of green employment is emerging across the globe and researchers are trying to find out option for green employment creation as an outcome of global climate change mitigation action (United Nations Environment Program, http://www.unep.org/civil-society/Implementation/GreenJobs/tabid/104810/Default.aspx). Since unemployment is like a chronic disease in the late-industrializing economies, policy thrust may go towards more employment generation. In this context, identifying sectors which have large impact on employment and less impact on GHG emissions would be a crucial policy challenge for the late-industrializing country like India. Therefore, understanding linkages between employment and GHG emissions is prerequisite for this analysis.

Now the changes in the above-mentioned exogenous variable may increase the employment opportunities within the domestic economy. Expectedly, pattern the employment opportunity will not increase uniformly across the sectors due to differences in labor intensity. To what extent the sectorwise employment will change is an important policy-relevant question. To answer this question, we have also derived employment multiplier from our SAM. The employment multiplier will show how the sectorwise employment changes due to any exogenous changes into the economy. So if we link this employment multiplier with the GHG emissions, we can show the impact of employment change on GHG emissions.

4.2 Method of Estimating SAM Multiplier

Before we estimate SAM multiplier, let us describe a SAM in a simple form for the understanding of multiplier analysis (Table 4.1).

Table 4.1 Schematic structure of social accounting matrix (SAM). (Source: Pradhan et al. (2006), concept and construction of SAM)

The four main sets of accounts, viz., factors, institutions, activities, and capital accounts are endogenous. All other accounts are collected together as fifth block in the schematic presentation in Table 4.1 and this is also exogenous account. They include the accounts for government account, indirect taxes, and combined current and capital accounts for rest of the world. In Table 4.1, T ₁₃ is a matrix of value added generated by the various production sectors, T ₃₃ gives the input–output transaction matrix. T ₂₁ maps the factor income distribution into the households income distribution and T ₃₂ reflects the expenditure pattern of the various institutions. Finally, T ₃₄ displays the investment demand for production activity while T ₄₂ shows the savings of the institutions. The later is also referred as leakages from the system.

The first step is to obtain matrices of coefficients of expenditure A _ij by dividing each element in the ij by corresponding sum of the column vector Y _j.

$$ A_{{\rm{ij}}}= T_{{\rm{ij}}} Y_{\rm{j}} ^{ - 1}\ldots\ldots {\rm{ }} = {\rm{ }}1,{\rm{ }}2,{\rm{ }}3,{\rm{ }}4 $$

(4.1)

Doing so implies that the accounting constraints across rows can be expressed as

$$ \left[ {\begin{array}{*{20}c} {Y_1 }\\[8pt] {Y_2 }\\[8pt] {Y_3 }\\[8pt] {Y_4 }\\[8pt] {Y_5 }\\[8pt]\end{array}} \right] = \left[ {\begin{array}{*{20}c} 0 & 0 & {A_{13} } & 0\\[8pt] {A_{21} } & 0 & 0 & 0\\[8pt] 0 & {A_{32} } & {A_{33} } & {A_{34} }\\[8pt] 0 & {A_{42} } & 0 & 0\\[8pt] 0 & {A_{52} } & {A_{53} } & 0\\[8pt]\end{array}} \right]\left[ {\begin{array}{*{20}c} {Y_1 }\\[8pt] {Y_2 }\\[8pt] {Y_3 }\\[8pt] {Y_4 }\\[8pt]\end{array}} \right] + \left[ {\begin{array}{*{20}c} {X_1 }\\[8pt] {X_2 }\\[8pt] {X_3 }\\[8pt] {X_4 }\\[8pt] {X_5 }\\[8pt]\end{array}} \right] $$

(4.2)

Where X _i is the row sums of sub-matrix T _{i . 5} for each i = 1, 2, 3, 4, 5. This is just another way of representing Table 4.1.

Subsequently, analysis assumes that each of the X _i’s is an exogenous set of numbers and that each of the A _ij matrices in equation has constant elements. Combining these two sets of assumptions implies that the values of Y ₁ to Y ₄ can always be obtained from any assumed values of X ₁ to X ₄ and can be expressed as

$$ \left[ {\begin{array}{*{20}c} {Y_1 }\\[8pt] {Y_2 }\\[8pt] {Y_3 }\\[8pt] {Y_4 }\\[8pt]\end{array}} \right] = \left[ {\begin{array}{*{20}c} 0 & 0 & {A_{13} } & 0\\[8pt] {A_{21} } & 0 & 0 & 0\\[8pt] 0 & {A_{32} } & {A_{33} } & {A_{34} }\\[8pt] 0 & {A_{42} } & 0 & 0\\[8pt]\end{array}} \right]\left[ {\begin{array}{*{20}c} {Y_1 }\\[8pt] {Y_2 }\\[8pt] {Y_3 }\\[8pt] {Y_4 }\\[8pt]\end{array}} \right] + \left[ {\begin{array}{*{20}c} {X_1 }\\[8pt] {X_2 }\\[8pt] {X_3 }\\[8pt] {X_4 }\\[8pt]\end{array}} \right] $$

(4.3)

$$ Y_5= A_{52} Y_2+ A_{53} Y_3+ X_5$$

(4.4)

The fifth set of accounts can be derived once Y ₂ and Y ₃ are known. That is, once the first three of accounts are balanced. The residual balance equation is not of further interest to us. Hence, our main analysis will be Eq. 4.3. This can be written as

$$ Y = AY + X $$

(4.5)

$$ Y = (I - A)^{ - 1} X $$

(4.6)

This shows that Y (i.e., Y ₁, Y ₂, Y ₃, and Y ₄) can be derived from X (i.e., X ₁, X ₂, X ₃, and X ₄) through a generalized inverse (I − A)⁻¹. This is analogous to that of conventional input–output analysis, which is concerned with the determination of Y ₃, the production activity accounts only. It is apparent that

$$ Y_3= ( {I - A_{33} } )^{ - 1} ( {A_{32} Y_2+ A_{34} Y_4+ X} ) $$

(4.7)

Equation 4.7 is a part of our system and, therefore, completely consistent with it. It is also the end of the story in the simplest form of input–output analysis since the latter assumes that A ₃₂ Y ₂ and A ₃₄ Y ₄ are exogenous. Thus, in this approach, Y ₃ (the level and structure of output) is derived through the inverse (I − A ₃₃)⁻¹ of direct and indirect commodity requirements on the basis of assumed demand on activities from other accounts. In Eq. 4.7, an obvious extension is to decompose these assumed demands and to allow for parts of them, A ₃₂ Y ₂ and A ₃₄ Y ₄ to be determined simultaneously with Y ₃. The later depends on Y ₂, i.e., on the level and distribution of income across institutions and on Y ₄, the saving investment balance.

Sometimes, the generalized inverse (I − A)⁻¹ is broken down into three matrices to reflect the different mechanism at work within it, resulting from the interconnections within the system. That is, the total multiplier of SAM is written as

$$ ( {I - A} )^{ - 1}= M = M_3 M_2 M_1$$

(4.8)

Where, the notation M is used to indicate a multiplier matrix. Note that M has rows and columns as factors (labor, capital), institutions (private, households, etc.), activities, and capital account. In this study, we have computed the M matrix on the basis of our 35 sector SAM of the year 2006−2007. This computed M matrix is shown in the Appendix 4.

The aggregate multiplier matrix, M, shows how an increase in any element of X _i will increase the corresponding element of Y _i by at least the same amount, and may have also indirect effects on other elements. The right-hand side of the equation shows that these aggregate multipliers can be decomposed into three separate multipliers. M ₁ and M ₂ are called as “own effects multipliers,” as opposed to M ₃ which collects together cross effects. The differences between M ₁ and M ₂ are as follows. A change in an element in the vector X _i of X will influence Y _i for two sets of reasons. One is that there may be transfers within the ith set of accounts so that, for example, an increase in demand on a production sector will cause it to increase its demand on another production sector. A second example is that increased income of companies from an exogenous source will result in increased income for government through profit taxation. Such multiplier processes which operate within a set of accounts can be referred to as “own direct effects” or “own transfer effects.” These contrast with M ₂ which collects together “own indirect effects” which arise from the fact that an increase in the elements of X _i will affect Y _i via other accounts. Thus, an increase in demand on a production activity will cause it to hire more factors. This will raise incomes in the factor accounts which in turn raise incomes in the institution accounts. These latter accounts will spend some of the increased income and in doing so will raise demand on the production account beyond the level of the initial increase which came from an exogenous source. Finally, multipliers M ₃ record cross effects, i.e., the impact of an increase in elements of X _i on Y _j for j ≠ i.

Though one can disaggregate the effects into three separate components, one needs for all practical purposes the total impact which can be obtained straightforward by working with M matrix. Thus, if our focus is on the impact on production activities due to increased spending by the household, one needs to look at the components of the matrix Mactivity × households.

So far, our discussion has said nothing about the consequence on environment of change through changes in exogenous sectors to the economy. This can be easily remedied by making assumption about the links between gross output and pollution generation in each production activity. The standard practice in this regard is to estimate direct pollution generation coefficients for each sector. These direct pollution generation coefficients are obtained from our ESAM by dividing column of each type of pollutants with the column of gross output (Table 3.4 of Chap. 3). The pollution generation coefficients thus obtained are expressed in terms of tons of pollution per lakhs of rupees of output. Once we have these coefficients, we can perform environmental impact analysis by estimating the pollution trade-off multipliers.

4.3 Pollution Trade-Off Multiplier

To estimate pollution trade-off multiplier, we have used method described by Robert (1975). The pollution trade-off multiplier measures the direct, indirect, and induced impact on pollution generation level due to exogenous change in sectoral output, households income, etc. The mathematical expression of the pollution trade-off multiplier is given as follows:

$$ E = P \cdot Y $$

(4.9)

where E is matrix of sectorwise emission, P is the sectorwise emission coefficient matrix.

Replacing Eq. 4.6 into Eq. 4.9,

$$ E = P \cdot ( {I - A} )^{ - 1}\cdot X $$

(4.10)

$$ \frac{{\partial E}}{{\partial X}} = P \cdot ( {I - A} )^{ - 1}= T $$

(4.11)

T is the pollution trade-off multiplier matrix which indicates impact on emission due to any exogenous changes into the economy. T matrix has different blocks like pollutant × activity block, pollutant × households block, etc. and these can be used for different types of impact analysis. For example, if we want to analyze impact of output growth on emission we have to use pollutants × activity block of the T matrix.

We have illustrated the methodology for estimating SAM multiplier as well as pollution trade-off multiplier. In what follows, we have used our ESAM to derive quantitative estimates which are relevant in the context of the objectives of this chapter.

4.4 Impact of Sectoral Output Growth Resulting from any Exogenous Changes in the Economy on Energy Use

Any production process consumes energy. So, any output increase would lead to incremental energy consumption unless there is any innovation in energy efficient technology. But, the innovation of energy efficient technology is not a quicker process. So, our objective in this study is to analyze the sectoral growth impact on energy use under the existing technological condition. Now this SAM multiplier model is a fixed technology static model. Therefore, the SAM multiplier model does not consider any technological improvement in the economy. Rather, it takes into account the direct, indirect, and induced effect on energy use under constant technological condition.

To analyze the total effect (i.e., direct, indirect-induced) on energy use, we have considered the energy × activity part of the SAM multiplier matrix. Our SAM takes into account seven types of energy commodities. However, we are concerned here mainly with the commercial energy available from primary conventional sources. The commercial energy commodities available from primary conventional sources are coal, crude oil, natural gas, hydro, and nuclear. Table 4.2 shows the impact of sectoral growth on this primary energy use due to any exogenous changes in the economy.

Table 4.2 Impact of sectoral growth on primary energy use for the production. (Source: Author’s estimation)

In this table, we have shown the direct and indirect-induced effect on energy use for all sectors except the above four types of primary energy sectors. We have also decomposed the total effect, i.e., direct, indirect-induced effect into two parts, viz., direct and indirect-induced effects. The direct effect on energy use is nothing but the direct energy coefficient, i.e., amount of energy to be used to produce one unit of output and this can be obtained directly from our SAM of the year 2006−2007. This is also called energy intensity with response to gross output of the sector. To estimate the indirect-induced effect, we have subtracted this direct effect from the total effects.

If we look at the impact on coal energy use for thermal electricity sector (NHY), we can see that the total effect on coal energy use is higher for the NHY sector (see Table 4.2). The total effect of NHY sector on coal is 0.2098 out of which 0.1131 is the direct and 0.0967 is the indirect-induced effect on coal energy use. This implies that if due to any exogenous changes the output of the NHY sector increases by one unit, the total coal energy requirement of the NHY sector will rise directly by the amount 0.1131 units and indirectly by the amount 0.0967 units.

Apart from the thermal electricity sector, the total impact on coal is observed higher for the iron and steel sector and the cement sector in comparison to other sectors of the economy (see Table 4.2). The total impacts on coal for these two sectors are 0.13 and 0.11, respectively, out of which the direct impacts are 0.07 and 0.06, respectively. Therefore, if the output of the iron and steel sector increases by one unit due to any exogenous changes, the economy wide demand for coal will be increased by 0.13 unit. The same logic is also valid for the cement sector.

By contrast, Table 4.2 shows that the direct and indirect-induced impact on coal is significantly low for agriculture and other services sector. The direct uses of coal in these sectors are negligible. Therefore, whatever increase in coal demand is observed for these sectors is due to indirect-induced effect.

In case of crude oil, we can observe from Table 4.2 that only the petroleum sector has significant direct impact on crude oil. This direct effect is negligible for some sectors and for most of the sectors this is almost zero. But when we take into account indirect-induced effect, the total impact on crude oil demand in the economy will increase significantly. For the thermal electricity sector and for most of the manufacturing sectors, we have observed significant impact on crude oil demand (see Table 4.2). We can say that though there are negligible direct impacts in some sectors but due to indirect-induced effect, their output growth will increase the economy wide demand for energy.

If we compare the direct and indirect-induced impact on different energy types, we can see this impact is higher for crude oil relative to other commercial energy type available from primary conventional sources (see Table 4.2). This impact is much lower in case of energy like hydro, natural gas, and nuclear. So, we can say that crude oil is the key source of primary energy in India.

However, the direct and indirect-induced impact on energy use and GHG emissions are observed due to the backward and forward linkage of the production sectors of the economy. The economic activity in a sector with high backward linkage provides stimulus to other sectors by requiring more inputs, whereas activity in a sector with high forward linkage stimulates higher outputs in other sectors by providing more inputs to them (Pradhan et al. 2006).

For illustration purpose, let us analyze the forward and backward linkage effect with an example of thermal electricity sector. The thermal electricity sector is a key source of the electric supply in case of Indian economy. The share of thermal electricity sector in total electric supply is about 86 % in the year 2006−2007.¹ So, the other production sectors of the economy are strongly dependent on the thermal electricity sector for their electricity requirement and hence the thermal electricity sector has a high forward linkage. Since there is high forward linkage, the expansion of thermal electricity sector will lead to expansion of the output of other sectors in the economy. Again the thermal electricity sector requires both energy and non-energy inputs for its production process and these are supplied by the other production sectors of the economy. If we look at the column of the thermal electricity sector from our environmentally extended SAM (ESAM), we can see that most of the nonenergy inputs of this sector are supplied from the energy-intensive sectors. Hence, total energy use in the energy-intensive sectors will increase due to expansion of thermal electricity sector.

Thus, we have demonstrated how the energy use will be increased if there is growth in sectoral output. Of course, this energy use causes GHG emissions in the atmosphere. According to Indian Network on Climate Change Assessment (INCCA) report, the energy-based GHG emissions is most in India. Almost 58 % of total carbon dioxide equivalent (CO₂EQ) emission is due to energy use (INCCA 2010). So our next objective is to analyze the impact on GHG emissions if the sectoral output grows due to any exogenous injection, and this is described in the following section.

4.5 Impact on GHG Emissions due to Sectoral Growth Resulting from any Exogenous Changes in the Economy

In this case, we have used the pollutant × activity block of pollution trade-off multiplier to analyze the impact of sectoral growth on the pollution emission, and this is given in Table 4.3. Each cell entry of Table 4.3 shows the direct and indirect-induced effect on generation of the pollutants due to change in the output of the production sectors. The higher value of multiplier in Table 4.3 implies higher impact on pollution generation. Therefore, the analysis would help us to find out the leading sectors of the economy, which have highest impact on the environment.

Table 4.3 Impact of output growth on greenhouse gas (GHG) emission (tons/lakhs of rupees of output). (Source: Author’s estimation)

As Table 4.3 shows, the direct, indirect-induced impact on CO₂EQ emission is highest (i.e., 86.88 t) for thermal electricity sector. Out of this total impact, the direct impact of thermal electricity sector is about 49.22 t (Table 4.3). Therefore, if the output of the thermal electricity sector is increased by one unit, the total CO₂EQ generation in India will rise by 86.88 t. Hence, the thermal electricity sector of India is the leading sector in terms of GHG emissions and the same thing is also evident from INCCA report (INCCA 2010). INCCA report also says that the CO₂EQ emission from thermal electricity sector is high due to coal use in its production process.w The coal constitutes about 90 % of the total fuel mix used in the thermal electricity sector. So, the increase in output in thermal electricity sector will have significant direct effect on total CO₂EQ emission in India. In the above section, we have discussed about the backward and forward linkage of thermal electricity sector. Here also, we have seen that backward and forward linkage affect significantly GHG emissions for thermal electricity sector.

After thermal electricity sector, the total impact on CO₂EQ emission is observed high for cement sector. As Table 4.3 shows, if the output of the cement sector increases by ₹ 1 lakh, the total CO₂EQ emission in India will increase by 53.88 t. The energy-based emission from the cement sector is almost 44 %, whereas production-process-based CO₂EQ emission is 56 % (INCCA 2010). Therefore, an increase in output of the cement sector will increase directly due to both energy use and production process. On the other hand, we have observed significant impact on coal use due to backward and forward linkage effect (see Table 4.2). Therefore, we have observed for the cement sector significant direct and indirect-induced impact on CO₂EQ emission.

On the other hand, the values of the direct pollution coefficients are very small for the agricultural sectors and other services sectors of the Indian economy (Table 4.3). If we look at the direct effect of paddy sector, we find that the CO₂EQ emission per unit of paddy output is about 6.29 t per ₹ 1 lakh of output. However, it is observed from Table 4.3 that the total effect on CO₂EQ emission generation is significant (i.e., 32.17) in paddy sector. In case of service sector, Table 4.3 indicates that the direct effect is negligible. But due to indirect-induced effect, the CO₂EQ emission in India will increase by 19.67 t resulting from an increase in service sector output by ₹ 1 lakh. Hence, we can conclude that those sectors that have small direct impact may have significant indirect and induced effect on the GHG emissions.

The above discussion indicates that the energy-intensive sector has the significant impact on energy use and GHG emissions. But it will be wrong to say that the growth in non energy-intensive sector will not have significant impact on energy use as well as on GHG emissions. This could happens because there are backward and forward linkage effects between the sector and the rest of the economy due to which there coould be significant impact on energy use and GHG emissions.

Till now, we have described the impact of output growth on GHG emissions in India. But this growth in output will increase the sectorwise employment in the economy. In the context of economic growth analysis in India, the analysis of employment change is essential. On the other hand, it will be interesting to see whether this increase in employment have adverse impact on GHG emissions in India or not. This issue is addressed in the following section.

4.6 Impact of Change in Employment on GHG Emissions

To analyze the impact of sector-specific employment change on GHG emissions, we have used method of employment multiplier. Before we estimate this employment multiplier, let us see the pattern of sectorwise employment for the year 2006−2007, and this is given in following Table 4.4.

Table 4.4 Sectorwise share of employment and labor intensity (2006−2007). (Source: www.indiastat.com and author’s estimation)

As this table shows, the share in total employment and labor intensity is high in the agriculture sector and other services sector. It is interesting to note that the share of employment as well as labor intensity is higher in other services sector in comparison to the manufacturing sector. Therefore, if sectoral output changes due to unitary changes in the exogenous account of SAM, the employment opportunity will increase more in agriculture and other services sector relative to manufacturing and other sectors of the economy. If we link GHG emissions with this employment growth, we can find the sector for which the impact of increase in GHG emissions is less as compared to other sector. To do this, we have first estimated sectorwise employment multiplier and then we relate that with the GHG emissions. The methodology and analysis is described below.

4.6.1 Employment Multiplier

Employment multiplier shows the direct and indirect changes in employment if the output of a sector changes due to any exogenous changes in the economy. Below, we have estimated the employment multiplier to show the direct and indirect-induced impact on employment due to any exogenous changes in the economy.

Let L _i be the employment in sector i and E _i be fixed employment coefficient.

Therefore,

$$ E_{\rm{i}} \,{\rm{ = }}\,L_{\rm{i}} {\rm{/}}Y_{\rm{i}}$$

(4.12)

Reorganizing this equation with Eq. 4.6 by substituting MX for Y, we may rewrite it as

$$ L = eMX $$

(4.13)

Where e is the diagonal matrix formed with elements E _i and the element of “e” corresponding to nonproduction account are zero.

Differentiating the above equation we get,

$$ \partial L/\partial X = eM = R $$

(4.14)

Therefore, R _i gives us the employment multiplier for sector i which indicates the direct, indirect, and induced employment created in the whole economy when the exogenous variable changes by one unit.

4.6.2 Impact of Employment Creation on GHG Emissions

To investigate the impact of employment creation on GHG emissions, we have to link this employment multiplier with the GHG emissions. To link GHG emission with employment multiplier, we have followed the same method as we have applied for pollution trade-off multiplier. Here, we have estimated sectorwise emission employment ratio by dividing sectorwise total emission from their corresponding level of employment (i.e., number of labor engaged in that particular sector). Premultiplying this with the employment multiplier, we have obtained the employment-emission multiplier and this is shown in the following equation.

$$ E_{\rm{P}}= e_{\rm{p}}\cdot R $$

(4.15)

where e _p is the employment emission ratio.

Thus, we have estimated the impact on sectoral GHG emissions due to increase in sectoral employment resulting from unitary changes in the exogenous account in our SAM. The result obtained in this way is given in Table 4.5.

Table 4.5 Impact of employment change on GHG emission. (Source: Author’s estimation)

From the above table, we can see that the value of employment multiplier is significant for agriculture and service sector relative to other sectors of the economy. This is due to their labor intensity as shown in Table 4.4. If we look at the impact on GHG emissions, we find that the increase in GHG emissions is not significant for service sector in comparison to the manufacturing and other sectors of the economy. In case of manufacturing sector, though its value of employment multiplier is less, it has significant impact on GHG emissions.

Thus, we have seen that the sectorwise employment will change at a different rate due to unitary changes in exogenous variable for all sectors. Also, we have seen that the impacts of employment change on GHG emissions are different for different sectors. This difference in impact is due to the sectorwise difference in energy consumption. In this case, it will be interesting to see the difference in energy consumption per unit of labor in the production sector. Here it is assumed that the energy consumption per unit of labor for the group of sectors will be followed by their counterpart. In Table 4.6, we have estimated energy consumption per unit of labor for broad group of sector in India for the year 2006−2007.

Table 4.6 Energy cost per unit of labor employed (energy in rupee value/labor). (Source: Author’s estimation with the help of SAM of the year 2006−2007)

It is clear from Table 4.6 that the payment for energy use per unit of labor is highest for energy production sector itself. In case of manufacturing sector, the payment for electricity use per unit of labor is almost 13 times higher than the other services sector. The payment for thermal electricity consumption is almost negligible for agriculture sector. Again in case of other forms of energy, the payment for energy use per unit of labor is significantly less than the manufacturing sector. As a result, despite the low employment multiplier the impact on GHG emissions is high for manufacturing sector as compared to the other services sector in India.

Furthermore, it can be observed from Table 4.7 and Fig. 4.1 that the share of other services sector in gross domestic product (GDP) is highest in India. In the year 2006−2007, the share of service sector is about 43 %, whereas the same for agriculture sector is 18 % and manufacturing sector is 15.59 %. On the other hand, the share of service sector is gradually increasing over the years, whereas this is declining for agricultural sector. In case of manufacturing, the share in GDP is almost constant during the period 1999−2000 to 2006−2007. Therefore, it is evident from Table 4.7 that Indian economy is structurally biased towards service sector. Hence, we can say that if the Indian economy follows this structure in future, the growth in service sector will boost economic growth, create significant employment opportunities with less impact on GHG emissions.

Fig. 4.1

Sector-specific shares in gross domestic product. (Source: CSO (2009), National Accounts Statistics)

Table 4.7 Sector-specific share in gross domestic product at factor cost. (Source: Author’s estimation with the help of national account statistics 2009)

Hence, this study shows that the impact on GHG emissions of the economy is not only dependent on the sector-specific energy intensity but also due to the backward and forward linkages between the activities. As a consequence of the linkage effect, the growth in output of low-energy-intensive sectors or low-emission-intensive sectors may have significant indirect-induced impact on total GHG emissions. On the other hand, if the objective is to create green employment in India, service sector growth will be the key as it has higher labor intensity with less GHG emissions intensity.

However, if backward and forward linkages between the sectors change, the indirect-induced impact may change, resulting in differential impact on GHG emissions. On the other hand, if any technological improvement changes the emission intensity, then direct impact on the GHG emissions may change. Now, this change in backward and forward linkage between the sectors will change if there is change in input–output coefficients. But the SAM multiplier model in this chapter is based on the fixed input–output coefficient of the production process. With this fixed input–output coefficient, it is not possible to do such kind of analysis. This needs further analysis, and we have taken into account this in the next chapter.