Decomposition of energy consumption and energy intensity in Indian manufacturing industries

Energy for Sustainable Development 14 (2010) 35–47

Contents lists available at ScienceDirect

Energy for Sustainable Development

Decomposition of energy consumption and energy intensity in Indian manufacturing industries B. Sudhakara Reddy ⁎, Binay Kumar Ray Indira Gandhi Institute of Development Research, Mumbai 400 065, India

a r t i c l e

i n f o

Article history: Received 27 April 2009 Revised 21 December 2009 Accepted 21 December 2009 Keywords: Efficiency Intensity Manufacturing sector

a b s t r a c t Of the total final energy consumption in India, the industrial sector accounts for about 36%, of which the manufacturing sector consumes about 66% (2004–2005 figures) with iron and steel, chemicals and petrochemicals, pulp and paper and cement industries being the largest energy users. In the recent past, energy intensity has been decreasing in the manufacturing sector, mainly due to fuel substitution away from coal in some sectors, most notably cement. Industrial output in developing countries like India continues to expand owing to rising populations and catching up on economic growth. This can result in higher energy use—energy provided primarily by the combustion of fossil fuels—and thereby to higher carbon dioxide (CO2) emissions. Using decomposition analysis, we show that most of the intensity reductions are driven purely by structural effect rather than actual improvement in energy efficiency. © 2009 International Energy Initiative. Published by Elsevier Inc. All rights reserved.

Introduction The oil crises of the 1970s led policy-makers to focus on the energy question, especially on the relationship between supply and demand, with particular emphasis on petroleum supply. It was recognized that the combustion of fossil fuels was the primary cause of climate change and this led to global discussions on energy supply and demand. A quantitative assessment of factors that contribute to changes in energy consumption has thus become imperative. Such analysis helps to understand the trends in energy use, to improve the efficiency of energy systems, to forecast future energy demand and carbon dioxide emissions and to measure the effectiveness of energy-related policies (Park, 1992; Farla et al., 1997). Since the industrial sector is a major consumer of energy, improvements in its energy efficiency are important to enhance productivity and reduce environmental impacts. Analysis of energy intensity indicators would give insights in this regard. In India, the industrial sector consumes about 36% of total energy and contributes about 30% of gross domestic product (GDP). Manufacturing accounts for two-thirds of this energy (CMIE, 2007). Improvement in the industrial energy intensity1 in the manufacturing sector leads to enhanced productivity of the whole economy and will improve overall environmental conditions. Major energy-intensive industries in India include: iron and steel, chemicals, textiles, aluminum, fertilizers, cement, paper and non-ferrous metals. Within ⁎ Corresponding author. Tel.: +91 22 28416526. E-mail address: [email protected] (B.S. Reddy). 1 The concept of industrial energy intensity denotes the amount of energy required to produce 1U of output.

the industrial sector, the main energy consumption operations are: driving motors (47%), electrical heating/melting (28%), air compression (10%) and lighting (4%). Different factors affecting the energy consumption for different processes and practices are technology design, age and sophistication of equipment, mechanical and chemical constraints, and external factors such as operating environment, maintenance and repair practices. These factors can be grouped under two broad categories—structural (related to organizational practices and environment) and technological (related to actual technologies in use). The question that arises then is how these two factors have affected energy intensity in India. To date, no studies have examined this dimension of energy use thus leaving a gap in India's energy research. Also, there exists an output factor, which simply symbolizes the increase in energy due to increase in industrial output. The decomposition of the overall change into these categories can provide policy-makers with the information needed to design appropriate strategies for reduction in energy use while helping mitigate environmental impacts. In the light of this, the present paper examines the usefulness of energy intensity indicators as policy tools in the context of issues related to the efficiency of India's industrial sector. The challenge is to reduce global GHG emissions without compromising the required energy services. The objectives of the present study are to: (i) analyze energy consumption and carbon dioxide emission for some selected energyintensive industries in the manufacturing sector, (ii) identify the factors that help reduce energy and carbon dioxide intensities, (iii) find the relative magnitudes of influence of each such factor, and (iv) draw policy prescription. The study follows the bottom-up approach to aggregate energy consumption, intensity, and emissions from micro level data. This implies that the data provided here are taken

0973-0826/$ – see front matter © 2009 International Energy Initiative. Published by Elsevier Inc. All rights reserved. doi:10.1016/j.esd.2009.12.001

36

B.S. Reddy, B.K. Ray / Energy for Sustainable Development 14 (2010) 35–47

from demand-side (energy consumption) instead of supply side. The correlation between the energy use and service output is investigated and the environmental impacts are assessed. This information has been used to develop energy as well as carbon dioxide intensity indicators for the most energy-intensive industries. Through the use of these indicators, governments may be able to identify the industries, and the processes and technologies to be targeted to improve energy efficiency and abate emission levels. Energy intensity and decomposition studies: literature overview Two basic approaches are used to express industrial energy intensity—i) physical energy intensity i.e. energy use per unit of physical output (e.g. PJ/tonne) and ii) economic energy intensity i.e. energy use per unit of monetary value of output (e.g. PJ/Rs billion).2 Physical intensities (energy and emission) are suitable for time series comparison of a single product, but are not preferred for the entire industrial sector as comparison of the units of physical output is difficult across sectors. Even in the same industry, various products may be produced which are valued differently, and may have different units. In such cases it is not possible to develop an aggregate measure of physical energy intensity. So, in this study the economic energy and emission intensities are used. The actual units of energy consumption and CO2 emission like kg, liter, cubic meter, Btu are converted to Joules and tonnes with the help of energy and emission conversion factors respectively. Energy intensity is expressed as TJ/Rs billion of output and emission intensity as thousand tonnes of CO2/Rs billion of output. It is worth noting that intensity per unit of economic output combines trends in intensity per unit of physical output with variations in the market price of the output produced.3 Energy intensity is inversely related to efficiency, which is given by energy consumption per unit of goods produced. The less energy required to produce a unit of output, the greater is the efficiency and lower the intensity. Three main factors affect the level of energy consumption in an economy\the level of overall activity or production, the composition or structure of the economy, and the output or activity per unit of energy consumed (Nooji et al., 2003). The change in total energy consumption between two periods of time is decomposed into these three separate components: (i) OE—output/activity effect (change in the production level), (ii) IE—‘pure’ energy intensity effect (accounting for change in technology, fuel shift/mix in the industries), and (iii) SE—structural effect (share of value of output change across the sector with in a given time) (Ang and Lee, 1994). This implies that the changes in energy consumption during the study period can be fully explained by activity, energy intensity and structural changes. The individual factor effect is estimated by keeping the other two effects invariant in the given time period. If energy intensity and structural effect are fixed at the base year, then the quantity of energy required to maintain the increased amount of activity levels is given only by activity level effect. In a similar way, the other two effects also can be defined. We need decomposition technique to isolate the different effects. The use of the decomposition method is not new to energy research as several past studies have used it (Jing et al., 1990; Boyd et al., 1988; Park, 1992; Lin and Chang, 1996). For a summary of various types of decomposition techniques, see Rose and Casler (1996). The method used here to compare and decompose energy consumption and energy intensity is the Divisia index model (difference and ratio). We use total differentiation model to calculate the changes. Choi and Ang (2003) have shown the symmetry 2

Rs stands for rupees—the Indian currency, the issuance of which is controlled by the Reserve Bank of India (RBI). The Rs and dollar conversion for the analysis period has been given in Appendix 1. 3 The analysis has been done at 1992 constant prices.

between “ratios method” and “differences method” of decomposition analysis and how with proper choice of formula these two alternative methodologies can be used almost interchangeably. Reddy (1998) has used the Divisia index approach to decompose Fiji's energy intensity change into different categories. According to De Bruyn (2000), if the input and output data were not available, some index numbers can be used for decomposition. Ang and Lee (1994) leave unexplained a major part of the observed changes in the energy consumption under decomposition. This means that the residuals give large estimation error in the decomposition analysis. Park (1992) has shown that the structural effect, calculated as a residual by the RRS technique (developed by Reitler et al., 1987), raises a number of logical questions. Firstly, the RRS method takes the mean between the base and the end periods for the static variables. The better alternative is to take values in the initial year.4 Secondly, and more importantly, the RRS method fails to introduce structural change explicitly as a variable in the equation. As a result, the RRS method may yield estimates at variance with those obtained from a method that incorporates the structural change variable. Hence, the structural effect, calculated as a residual by the RRS method is overestimated as it contains more than the effect of variables other than structural change. Sun (1998) has used a complete decomposition model where residuals are decomposed by the jointly created and equally distributed rule and compared the results with the general decomposition modeling. Bhattacharya and Paul (2001) used the total decomposition approach on energy consumption and energy intensity at sectoral level like agriculture, industry and transport. They have shown that the intensity effect contributes significantly to energy conservation. However, they have not disaggregated the analysis at industry level and for CO2 emissions and intensity. In the present analysis, we use the “Sun method” of total decomposition analysis. We decompose the energy consumption, energy intensity, CO2 emission and CO2 intensity at industry level over the years. Methodology of the study There are many methods to estimate and isolate the effects of different factors and create an intensity index. All the methods deal with creating the index for pure energy intensity by decomposing energy consumption and energy intensity. Here total differential decomposition model is used since energy content varies across fuels and the substitution among energy carriers affects total energy consumption. Edwards and Parikh (1978) had shown that the possibilities of substitution between fossil fuels on one hand and electricity on the other are less than those among fossil fuels, but there are certain complementarities. Hence, a more logically consistent method has to be formulated for decomposing energy consumption and energy intensity. The study estimates and evaluates economic energy intensity indicators using decomposition analysis at industry level. The analysis utilizes an empirical method, as explained in the literature review, to examine the factors, structural, and technological changes. Comparisons of energy intensities— among industry or countries or against “best practices” benchmark— can indicate opportunities for improvements in energy and process efficiencies. Because of inter-industry comparison, in the present study, we have used economic output indicator to measure the intensity. In this study, we first analyze the energy consumption of the Indian industry and assess the increases in energy use that come

4 For instance, the structural effect on industrial energy consumption between any two periods can be derived by measuring the change in energy consumption associated with a change in the industrial composition during the period. This can be done while holding all other variables—energy intensities of the individual industry branches and total industrial output—of the base period constant at their initial values.

B.S. Reddy, B.K. Ray / Energy for Sustainable Development 14 (2010) 35–47

with growth in output, and the environmental impacts that accompany the increase in fuel consumption. The energy and carbon dioxide intensities are found for the most energy-intensive industries, viz., iron and steel, aluminum, copper, textile, pulp and paper, cement, and fertilizer. The period of analysis is from 1992 to 2005. The trends in carbon dioxide intensity and the major factors that affect it (structural and economic changes) provide climate change policy-makers with the wherewithal needed to set CO2 targets for various industries, as well as design appropriate CO2 abatement strategies. Hence, these serve not just as monitoring tools, but also as a basis for energy efficiency policies and regulations aimed at achieving greater energy conservation. As a result, these indicators, particularly cross-country comparisons of them, are increasingly being touted as very useful and necessary instruments for policy making. To develop economic intensity indicators, energy consumption and physical production data have been collected at firm and specific energy consumption level and from secondary sources—Center for Monitoring Indian Economy (CMIE) (Prowess) and the energy profiles published by it. Thirteen types of manufacturing industries have been selected from the CMIE database. They are5: (1) chemical (fertilizer, inorganic chemicals, organic chemicals, pesticide, cosmetics, plastic product, polymers, tire and tube), (2) beverages and tobacco, (3) food products, (4) machinery, (5) aluminum and aluminum products, (6) copper and copper products, (7) iron and steel, (8) mining, (9) cement, (10) other nonmetallic and mineral products, (11) paper and paper products, (12) textiles, and (13) transport equipment. The industries and sub-industries included in the analysis are determined solely on the basis of the availability and quality of data. The data on energy use have been obtained for energy carriers grouped into three categories6: (i) solid fuel (ii) electricity and (iii) petroleum products and gaseous fuels. The constituents of the above categories are as follows: • Solid fuel: coal, coke, bagasse, firewood, etc.; • Electricity: electricity purchased; and • Petroleum and gaseous products: fuel oil, furnace oil, high speed diesel (HSD) and light diesel oil (LDO), low sulphur heavy stock (LSHS), and others. Firm-level total energy consumption is obtained by adding the product of all specific energy use with corresponding energy conversion values. The energy consumption data have been aggregated initially at firm level and then at industry level by combining energy consumption of all firms for a given industry type. Similarly, CO2 emissions are calculated by multiplying the energy use with the corresponding emission factors. The data taken here are at the micro level and hence are more reliable than the macro level data as very often these data are taken from the supply side of the energy of a sector and are not true representatives of the actual consumption level at the firm and industry levels (due to leakage, stocking, etc.). Total decomposition analysis model The approach applied for decomposing change in total energy consumption and energy intensity in the manufacturing sector in the Indian industry during1992–2005 is a simple total differential

method. In this technique, the residual obtained is due to combined effect distributed among the output effect, intensity and structural effect. The manufacturing sector energy intensity and product mix effects are taken as major components for decomposition of the manufacturing sector change in energy intensity. Decompositions of total energy consumption and total carbon dioxide emissions are given as: Et =

m X

Eit =

i=1

m X k X

ð1Þ

Eijt

i=1 j=1

Total energy consumption is the summation of all types of energy consumed by any sector or firm to produce the output. In Eq. (1), j denotes the types of energy consumed and i the industrial sub-sector. We can multiply and divide the output on right-hand side in such a way that the original equation (Eq. (1)) would be intact. Et =

m X Eit P T m it P P it i=1 i=1

Et =

m X

T Pit

m X

ð2Þ

Pit

i=1

eit Tα it Tpt

ð3Þ

i=1

where Eit = energy consumption or carbon dioxide emission of industry i at time t Pit = total value of output of industry i at time t eit = EPitit = energy or carbon dioxide emission intensity of industry i at time t Pit α it = P = share of value of output of industry i at time t and m Pit

i=1

Pt = total value of output at time t. The total decomposition in different factors can be given as ΔEt = ΔOt + Δet + ΔSTt m X

Δet =

ΔOt =

+ 1 = 3ðΔPt · Δα it ÞÞ

m X

m X

ð5Þ

ΔPt ððeit · α it Þ + 1 = 2ðΔeit · α it + eit · Δα it Þ

i=1

Δt =

ð4Þ

Δeit ððPt · α it Þ + 1 = 2ðΔPt · α it + Pt · Δα it Þ

i=1

+ 1 = 3ðΔeit · Δα it ÞÞ

ð6Þ

Δα it ððPt · eit Þ + 1 = 2ðΔPt · eit + Pt · Δeit Þ

i=1

+ 1 = 3ðΔPt · Δeit ÞÞ

ð7Þ

where ΔOt = change in energy consumption or carbon dioxide emission due to output effect Δet = change in energy consumption or carbon dioxide emission due to energy intensity effect ΔSTt = change in energy consumption or carbon dioxide emission due to structural effect and ΔEt = change in total consumption. Decomposition of energy intensity and carbon dioxide intensity: m k P P

5

This classification is based on the National Industrial Classification (NIC), which is the standard followed for categorising economic activities. The NIC is prepared to suit the Indian conditions and follows the principles and procedures laid down in the United Nations' International Standard Industrial Classification (ISIC) of all economic activities. The structure of NIC-2004 is identical to the structure of ISIC Rev.3.1 up to 4digit level class (GoI, 2008). 6 Strictly there are close to 40 categories of fuel used in the industry. For representative purpose, fuels have been clubbed into three groups. For each group, the calculation is done by taking the calorific values of each constituent of the group.

37

et =

i=1 j=1 m P i=1

et =

m P

Eijt

Pit

=

i=1 m P i=1

Eit = Pit

m m X X Eit Pit T = eir Tα it : P p t i = 1 it i=1

Et Pt

ð8Þ

ð9Þ

38

B.S. Reddy, B.K. Ray / Energy for Sustainable Development 14 (2010) 35–47

Table 1 Total primary and commercial energy consumption in India (1970–2000). Energy use (PJ or %)

GPRAa (%)

Year

Primary energy consumption Commercial energy consumption Commercial share, % Commercial energy consumption by industry Industry share in commercial energy consumption, %

1970–71

1980–81

1990–91

2000–01

2004–05

6274 2488 39.7 1235 49.64

8884 3440 38.7 1418 41.22

13017 6721 51.6 2609 38.82

18668 9735 52.2 3596 36.94

22040 11852 58 4283 36

3.7 4.6 3.6

Source: CMIE Reports. Note: primary energy is the energy embodied in natural resources prior to undergoing any human-made conversions or transformations. For instance, in the case of coal based thermal power plant, primary energy would refer to the heat content of coal. Similarly, for hydro-electricity the primary energy is the potential energy associated with water level. Commercial energy refers to the sources of energy which command a price and are subjected to commercial exchange. Examples of commercial energy are coal, petroleum, and electricity. On the contrary, non-commercial sources of energy are supposed to be free and consist of firewood, vegetable wastes, agri-waste, animal-waste etc. Commercial sources are mostly exhaustible (with hydroelectric power being an exception), and easily quantifiable, whereas non-commercial sources of energy are mostly renewable and difficult to account for in energy balances. a GRPA = growth rate per annum.

The manufacturing sector energy intensity and product mix effect are taken as major components for decomposition of the manufacturing sector change in energy intensity. Taking total differentiation of previous intensity equation w.r.t. time t

t

t

Δetotal = ΔeIE + ΔeSE t

ΔeIE =

m X

ð10Þ

Δeit ðα it + 1 = 2ðΔα it ÞÞ

ð11Þ

Δα it ðeit + 1 = 2ðΔeit ÞÞ:

ð12Þ

i=1

t

ΔeSE =

m X i=1

where ΔetIE = change due to technical or pure intensity effect at time t ΔetSE = change due to structural effect at time t Δettotal = change in total energy intensity at time t. Pattern of energy consumption and output in Indian industry Energy use In India, between 1970–1971 and 2004–2005, the primary energy consumption has increased from 6274 to 22,040 PJ, at a compound growth rate of 3.7% per annum (Table 1). During the same period, commercial energy use grew at a higher rate of 4.66% per annum. The reasons for the growth of commercial energy more than primary energy are: (i) substitution of non-commercial energy with commercial energy and (ii) increase in commercial energy use due to changing lifestyles (Reddy, 2004). The share of commercial energy use in industry is more than any other sector and is followed by transport, household, agriculture and commercial sectors (Table 2). Even though

industry is the main consumer of commercial energy, its share in the total decreased from 41.3% in 1960–1961 to 36.5% during 2004–2005 at a rate of 1% per annum. In the industrial sector, solid fuels have the highest share among all fuels (Table 3). Although consumption increased over the study period, they have lost share to natural gas. With a compound growth rate of 13.4% per year, natural gas increased its share of total energy from 2.7% in 1980 to 18.0% in 2005. The consumption of petroleum products has been around 20.3% and that of electricity around 11.5%. The four major manufacturing sub-sectors which use a sizable share of overall energy are chemicals and petrochemicals, iron and steel, pulp and paper and cement. In the manufacturing sector, heat production is one of the most important uses of energy. For end uses such as heat treatment, melting and smelting, and cement calcinations, direct heat and steam are used. Other important end uses are machine drive and electrolytic processing. There are other uses such as ventilation, airconditioning, lighting of industrial facilities and on-site transport. Variations in energy use in the industrial sector result from technical and process changes that affect industrial sector activity, energy use, and emissions. Changes are made in the fuel types used by industries in response to change in technologies, economic situations, or environmental regulations. For the present study, we analyze 13 sub-sectors that include iron and steel, chemicals, pulp and paper, cement, etc. accounting for about 66% total industrial energy use in 2005. Table 4 shows the disaggregate consumption in total manufacturing by each category. During 1992–2005, total industrial consumption has increased by 4.76% per annum. This is due largely to increase consumption of energy in textiles, chemicals and paper, which have grown by 16.4, 7.7, and 7.9% per annum, respectively. The largest percent increase per annum is in textiles, jumping 16.4% since 1992 as the industry has shifted from manual operations to automation and mechanization. The shares of iron and steel, and

Table 2 Sector wise commercial energy consumption in India (PJ). Sector

Industry Transport Household Services Agriculture Total

Year

GPRA (%)

1960–61

1970–71

1980–81

1990–91

2000–01

2004–05

685 (41.3) 664 (40.2) 160 (9.7) 116 (7.0) 27 (1.6) 1652 (100.0)

1235 701 304 181 67 2488

1418 943 389 437 253 3440

2609 (38.8) 1473 (21.9) 878 (13.1) 1153 (17.2) 610 (9.1) 6721 (100.0)

3596 2501 1248 1941 625 9911

4327 (36.5) 2855 (23.4) 1532 (12.5) 2620 (21.5) 855 (7.0) 12180 (100.0)

Source: CMIE, 2007. Note: figures in parentheses represent percentages.

(49.6) (28.2) (12.2) (7.3) (2.7) (100.0)

(41.2) (27.4) (11.3) (12.7) (7.4) (100.0)

(36.3) (25.2) (12.6) (19.6) (6.3) (100.0)

4.18 3.29 5.15 7.17 7.98 4.43

B.S. Reddy, B.K. Ray / Energy for Sustainable Development 14 (2010) 35–47

39

Table 3 Utilization of various energy carriers in industry—PJ (1980–2005). Year

Coal

%share

Electricitya

% share

Petroleum productsb

% share

Gas

% share

Total

1980–1981 1981–1982 1982–1983 1983–1984 1984–1985 1985–1986 1986–1987 1987–1988 1988–1989 1989–1990 1990–1991 1991–1992 1992–1993 1993–1994 1994–1995 1995–1996 1996–1997 1997–1998 1998–1999 1999–2000 2000–2001 2001–2002 2002–2003 2003–2004 2004–2005

892 1071 1150 1224 1244 1256 1272 1338 1501 1484 1648 1767 1815 1860 1824 1862 1871 1872 1818 1686 1708 2032 2078 2123 2169

62.9 63.2 64.6 77.6 63.3 61.4 59.2 60.4 59.7 57.8 57.9 58.6 58.7 57.6 56.5 55.5 54.1 52.9 51.3 47.9 47.5 51.1 50.8 50.4 50.1

176 195 194 209 231 246 258 254 277 296 337 353 346 371 374 406 405 407 408 405 414 458 472 485 498

12.4 11.5 10.9 13.3 11.7 12 12 11.5 11 11.5 11.8 11.7 11.2 11.5 11.6 12.1 11.7 11.5 11.5 11.5 11.5 11.5 11.5 11.5 11.5

313 371 367 65 395 412 440 430 467 454 501 519 507 546 575 634 712 729 769 778 780 802 828 854 881

22 21.9 20.6 4.1 20.1 20.2 20.4 19.4 18.5 17.7 17.6 17.2 16.4 16.9 17.8 18.9 20.6 20.6 21.7 22.1 21.7 20.2 20.2 20.3 20.3

38 58 69 79 96 132 180 192 272 334 358 377 424 452 452 449 470 531 549 651 694 683 715 748 780

2.7 3.4 3.9 5.0 4.9 6.5 8.4 8.7 10.8 13 12.6 12.5 13.7 14 14 13.4 13.6 15 15.5 18.5 19.3 17.2 17.5 17.8 18.0

1419 1695 1780 1577 1966 2046 2150 2214 2517 2568 2844 3016 3091 3229 3225 3351 3458 3538 3544 3519 3596 3975 4092 4210 4327

Source: CMIE, 2007. a Electricity, though secondary is considered here as an input energy to industry like primary energies such as coal, gas and petroleum products. b Petroleum products refer to materials derived from petroleum, which include diesel, light fuel oil, heavy fuel oil, petroleum coke, etc.

cement in total energy consumption in the year 2005 declined by 13 and 3%, respectively from that in 1992. Other manufacturing subsectors which includes services, electronic and telecommunication, construction, plastic, shipping, film, food, leather, apparel, gem and jeweler, cosmetic, etc., consumed 34% of total energy during the year 2005. Pulp and paper consume large amounts of energy in the form of biomass fuels, namely spent pulping liquor and solid wood waste. Over the study period, the shares of energy consumption in other categories decreased from 49 to 34% between 1992 and 2005 while that in chemicals increased by 7.7 percentage points per annum. Consumption in the petroleum refining industry dropped significantly. The specific composition of energy use in the industrial sector is shown in Table 5. In 1992, final energy consumption was dominated by solid fuels with 75.3%, which has declined to 57.2% in year 2005. Table 4 Energy consumption by various categories of industries (PJ). Sector Aluminum Beverages and tobacco Cement Chemicals Copper Food products Iron and steel Machinery Mining Other nonmetals Paper Textiles Transport equipment Total Others Grand total

1992

%total manufacturing

1995

2000

2005

%total manufacturing

185 2

12.8 0.1

202 3

260 3

351 4

12.2 0.1

275 231 5 15 562 1 1 32 88 47 2

19.0 16.0 0.3 1.1 38.9 0.1 0.0 2.2 6.1 3.3 0.2

236 287 5 22 595 2 1 42 121 65 4

315 463 11 46 702 3 5 38 181 245 4

471 605 13 57 733 3 6 43 238 339 4

16.4 21.1 0.4 2.0 25.6 0.1 0.2 1.5 8.3 11.8 0.1

1445 1399 2844

100.0

1583 1642 3225

2274 1245 3519

2867 1461 4327

100.0

The overall specific energy consumption increase in electricity, petrol and electricity sectors are: from 8.37 to 21 and 9.7 to 15.2% during the study period. Table 6 shows the value of output in various categories of industry. The main argument for using value-added as an output measure is that it relates more closely to the productive activity of the plant. In a stable macro- and microeconomic climate, such a measurement could be safely used as an indicator of production and a valid energy intensity estimate could be derived. The production of the chemical industry increased by 9.7% per annum, textiles by 7%, paper products by 8.3%, cement industries by 4.6%, copper by 9.7% and aluminum industries by 5.3%. Between 1992 and 2005, energy use increased at a rate of 5.4% per annum compounded, while the value of output has increased from Rs 1797 to 5308 billion at a rate of 8.7% per annum. The growth rate of output outstrips that of commercial energy use which means that there is an overall decrease in energy intensity in the manufacturing sector. A notable feature of the industries is that with modernization there is a trend towards declining energy use per unit of output measured at 1992 constant prices (Table 7). Taking into account the variations in the quality of domestic fuels and the profitability of the commodities produced by each industry, the data show that energy intensities appear to have substantial room for improvement. Carbon dioxide emissions Since energy consumption is responsible for roughly 90% of CO2 emissions in India (IEA, 2007), it is important to estimate emission intensity indicators which can be used for environmental monitoring. Energy efficiency improvements are an important tool for mitigating greenhouse gas emissions. In this study, carbon dioxide emissions from industry are calculated for all categories of industries.7 The

7 The CO2 emissions are based on only the carbon content in different types of fuels and process-related emissions are also considered.

40

B.S. Reddy, B.K. Ray / Energy for Sustainable Development 14 (2010) 35–47

Table 5 Energy consumption—carrier-wise in PJ (1992 and 2005). Sector

Aluminum Beverage and tobacco Cement Chemicals Copper Food products Iron and steel Machinery Mining Other nonmetals Paper Textiles Transport equipment Total % share

1992

2005

Coal

Electricity

Gas

Petroleum products

Total

Coal

Electricity

Gas

Petroleum products

Total

166 1 242 59 0 2 525 0 0 3 67 23 0 1089 75.31

13 0 20 31 0 1 18 1 0 25 15 15 1 140 9.68

0 0 8 76 5 1 0 0 0 1 4 0 0 96 6.64

5 0 5 65 0 11 18 1 0 3 3 9 2 121 8.37

185 2 275 231 5 15 561 1 1 32 88 47 2 1446 100.00

255 2 392 112 0 4 645 0 1 3 205 18 0 1639 57.15

43 1 52 81 1 8 46 2 3 19 22 154 2 435 15.17

0 0 5 159 5 2 2 0 0 9 4 4 0 190 6.62

53 0 22 252 7 43 40 1 2 13 7 162 2 604 21.06

351 4 471 605 13 57 734 3 6 44 238 339 4 2868 100.00

Source: CMIE (Prowess).

Table 6 Industry output by various categories—Rs billion (1992–2005) (base price 1991–1992). Industry

Aluminum Beverages and tobacco Cement Chemicals Copper Food products Iron and steel Machinery Mining Other nonmetals Paper Textiles Transport equipment Total

Output (billion's Rs) 1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

GR PA

27 32 61 725 13 68 193 244 101 20 28 135 150 1797

28 35 56 769 13 74 203 253 106 22 30 141 141 1871

28 39 56 780 15 85 207 270 114 28 32 161 155 1970

31 43 61 908 14 100 228 305 132 32 34 187 188 2263

37 41 69 997 18 124 263 357 138 36 40 221 236 2577

38 43 69 1175 21 147 292 397 156 41 40 245 281 2945

41 47 68 1164 20 177 281 420 175 49 44 266 268 3020

40 48 68 1379 25 203 267 449 160 51 41 265 264 3260

44 51 70 1710 33 214 278 508 207 58 43 284 317 3817

47 57 77 2001 38 222 311 566 231 66 54 298 325 4293

46 64 82 1900 34 226 308 600 229 69 57 268 340 4223

48 68 92 2061 37 250 320 647 245 77 65 284 364 4560

50 72 102 2236 40 277 333 699 263 85 72 301 391 4921

53 76 110 2426 43 307 347 754 282 94 79 319 418 5309

5.27 6.91 4.64 9.74 9.72 12.28 4.61 9.07 8.22 12.69 8.31 6.84 8.21 8.69

has now emerged as the leader in terms of carbon dioxide emissions which increased 8.1 times. Emissions from food, beverage and mining have grown even more. Of all of the industrial sub-sectors, only nonmetals show a decline. Overall, the pattern of emissions shown in the table reinforces the message that emissions from the industrial sector have increased substantially, which makes a strong case for adopting energy-efficient measures.

calculations are based on the emission conversion factor of the European Topic Centre for Air and Climate Change ETC/ACC (2003) and Earth Trends (2003) given in Appendix 2. Table 8 shows that emissions from both steel and cement industries have grown nearly 1.4 and 1.8 times, respectively, during the study period, while those from copper, chemical and textile industries have more than doubled. Interestingly, the textile industry

Table 7 Energy intensity in various categories of industries (TJ/Rs billion) (base price 1991–1992). Industry

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

GR PA

Aluminum Beverage and tobacco Cement Chemicals Copper Food products Iron and steel Machinery Mining Other nonmetals Paper Textiles Transport equipment Total

6837 56 4502 318 369 225 2909 5 6 1605 3148 350 16 804

6711 71 3989 338 362 223 2638 6 6 1709 3101 458 18 764

6843 74 4005 323 367 222 2636 6 5 1400 3245 342 19 733

6500 67 3870 316 350 221 2609 6 6 1316 3552 345 20 699

5565 80 3627 301 344 222 2411 6 6 1197 3229 332 17 652

5968 60 3914 281 338 213 2352 6 6 915 3264 312 16 614

5571 57 4318 299 360 216 2453 6 5 741 3406 468 15 638

6108 54 4464 276 336 204 2612 6 20 706 3786 534 14 620

5900 49 4500 271 327 213 2525 5 24 657 4199 862 14 596

5632 37 3919 270 324 217 2033 6 646 558 3729 842 11 570

6585 55 4089 275 321 216 2056 5 21 542 3560 1091 10 568

6611 54 3878 266 310 205 2075 5 21 513 3305 1081 10 554

6637 54 4228 257 301 194 2095 5 21 485 3147 1072 10 551

6663 53 4285 249 291 184 2115 4 20 459 3014 1062 9 540

− 0.20 − 0.44 − 0.38 − 1.87 − 1.82 − 1.52 − 2.42 − 1.37 9.85 − 9.19 − 0.33 8.92 − 4.00 − 3.02

B.S. Reddy, B.K. Ray / Energy for Sustainable Development 14 (2010) 35–47

41

Table 8 Carbon dioxide emissions (Mt). 1992

Aluminum Beverage and tobacco Cement Chemical Copper Food product Iron and steel Machinery Mining Other nonmetals Paper Textiles Transport Total

2005

Coal

Electricity

Gas

Petroleum

Total

Coal

Electricity

Gas

Petroleum

Total

18.46 0.14 26.88 6.52 0.00 0.67 58.24 0.01 0.05 0.34 7.41 2.59 0.00 121.31

4.11 0.09 6.47 9.85 0.01 1.06 5.83 0.17 0.01 7.87 4.59 4.62 0.19 44.88

0.00 0.00 0.52 5.04 0.33 0.08 0.01 0.00 0.00 0.07 0.02 0.03 0.01 6.11

0.41 0.02 0.35 5.23 0.00 2.06 1.40 0.05 0.01 0.23 0.21 0.67 0.12 10.76

22.98 0.25 34.22 26.63 0.35 3.86 65.49 0.23 0.07 8.52 12.24 7.91 0.33 183.07

28.29 0.25 43.52 12.46 0.00 0.91 71.60 0.04 0.06 0.38 22.79 2.03 0.01 182.35

13.65 0.41 16.50 25.72 0.30 4.74 14.42 0.64 1.04 6.08 6.96 48.69 0.56 139.72

0.00 0.02 0.31 10.50 0.32 0.24 0.16 0.00 0.00 0.56 0.25 0.29 0.01 12.66

4.07 0.02 1.72 19.93 0.54 6.59 3.11 0.07 0.14 0.97 0.54 12.49 0.15 50.32

46.01 0.70 62.05 68.61 1.16 12.48 89.29 0.75 1.24 7.98 30.55 63.50 0.73 385.06

GRPA (%) 5.49 8.24 4.68 7.55 9.76 9.44 2.41 9.40 24.54 −0.50 7.29 17.37 6.41 5.89

Data and source: given in Appendix 2.

Table 9 Carbon dioxide emission intensities (1000 t CO2/billion's Rs of output). Industry

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

GR PA

Aluminum Beverage and tobacco Cement Chemicals Copper Food products Iron and steel Machinery Mining Other nonmetals Paper Textiles Transport equipment Total

851 8 561 37 27 57 339 1 1 426 437 59 2 102

971 9 503 40 25 60 309 1 1 437 455 81 3 101

974 9 520 39 27 50 312 1 1 349 476 64 3 98

903 9 504 37 32 57 309 1 1 322 514 62 3 93

774 10 467 34 32 54 288 1 1 276 447 59 3 85

823 7 506 33 25 48 282 1 1 198 451 55 2 80

770 6 557 35 27 41 285 1 1 159 447 103 2 84

844 6 584 33 32 40 300 1 4 148 507 126 2 83

817 9 593 30 31 38 294 1 4 126 553 159 2 78

777 5 507 29 29 43 244 1 51 107 482 155 2 73

863 9 538 31 29 48 250 1 5 100 457 205 2 76

852 9 521 28 25 40 270 1 2 109 419 138 2 70

863 9 568 28 26 40 263 1 3 96 400 166 2 72

874 9 564 28 27 41 257 1 4 84 387 199 2 73

0.20 1.24 0.04 − 1.99 0.04 − 2.53 − 2.10 0.31 15.09 − 11.70 − 0.94 9.86 − 1.67 − 2.58

Table 9 provides carbon dioxide intensities among various industrial sub-sectors. The intensity in some sub-sectors increased substantially during the period shown, others have decreased, yet some others have risen and then fallen over time. The intensities are defined per unit of economic output (given in the value of the goods produced). Since they are expressed in economic terms, extra care must therefore be taken while interpreting carbon dioxide intensity trends, particularly when comparing results from categories of industries.

As the industrial sector expanded during the study period, energy consumption as well as carbon dioxide emissions increased. An important observation is evident when one considers the proportion of energy use and emissions contributed by the industry. Table 10 provides information on energy consumption, value of output, intensities of energy and CO2 emissions for 1992–2005 at an aggregate level. The real value of output (Rs billion) increased by 2.95 times, while energy consumption (PJ) increased only by 1.98 times. This shows an increase in energy efficiency and reduction in

Table 10 Energy consumption, output, energy and CO2 intensities. Year

Total energy consumption (PJ)

CO2 emission (MT)

Value of output (Rs billion)

Energy intensity (TJ/Rs billion)

CO2 intensity (′000 tCO2/Rs billion)

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005

1445 1430 1444 1583 1679 1807 1925 2021 2273 2447 2400 2527 2712 2867

183 189 192 209 220 236 255 272 298 312 323 321 354 385

1797 1871 1970 2263 2577 2945 3020 3260 3817 4293 4223 4560 4921 5309

804 764 733 699 652 614 638 620 596 570 568 554 551 540

101.9 101.2 97.7 92.6 85.3 80.1 84.5 83.5 78.1 72.8 76.4 70.4 71.9 72.5

42

B.S. Reddy, B.K. Ray / Energy for Sustainable Development 14 (2010) 35–47

Table 11 Decomposition of change in energy consumption sector wise (PJ) (1992–2005). Industry

Aluminum Beverage and tobacco Cement Chemicals Copper Food products Iron and steel Machinery Mining Other nonmetals Paper Textiles Transport equipment Total

Actual value

Shares in total industry %

Shares in total effect %

OE

IE

SE

Total

OE

IE

SE

Total

OE

IE

SE

295.9 3.1 422.6 427.7 8.9 34.1 771.7 2.4 2.5 50.2 164.8 164.4 3.6 2352.0

− 7.4 − 0.2 − 20.2 − 106.8 − 2.2 − 7.1 − 234.0 − 0.4 2.7 − 61.2 − 7.2 168.0 − 1.9 − 277.9

− 122.2 − 0.7 − 205.6 52.9 1.1 14.2 − 365.8 0.1 − 0.2 22.2 − 7.6 − 40.9 − 0.2 − 652.5

166.3 2.3 196.7 373.8 7.8 41.2 171.9 2.1 5.1 11.2 150.0 291.5 1.5 1421.5

12.6 0.1 18.0 18.2 0.4 1.5 32.8 0.1 0.1 2.1 7.0 7.0 0.2 100.0

2.7 0.1 7.3 38.4 0.8 2.6 84.2 0.2 − 1.0 22.0 2.6 − 60.4 0.7 100.0

18.7 0.1 31.5 − 8.1 − 0.2 − 2.2 56.1 0.0 0.0 − 3.4 1.2 6.3 0.0 100.0

11.7 0.2 13.8 26.3 0.6 2.9 12.1 0.1 0.4 0.8 10.5 20.5 0.1 100.0

177.9 137.2 214.8 114.4 113.8 82.8 448.8 115.8 49.0 446.9 109.9 56.4 235.9 165.5

− 4.5 − 7.8 − 10.3 − 28.6 − 27.6 − 17.3 − 136.1 − 21.0 54.0 − 544.7 − 4.8 57.6 − 122.9 − 19.6

− 73.5 − 29.4 − 104.5 14.2 13.8 34.5 − 212.7 5.2 − 3.1 197.8 − 5.1 − 14.0 − 13.0 − 45.9

Note. OE = output effect; IE = intensity effect; and SE = structural effect.

energy intensity which moved from 804 to 540 TJ/Rs billion. The overall CO2 emission intensity has declined from 102 to 72.5 ′000 tCO2/Rs billion. Based on the available data the energy requirement in the year 2005 can be estimated to be 4269 PJ if there were no changes in the energy intensity8 and structure9. But the total energy requirement in year 2005 was only 2867 PJ. This shows that there was an increase in output without a proportional increase in energy use because of structural shift or increase in energy efficiency or both. Similarly, it can be shown that CO2 emission for the year 2005 would have been 541 million tonnes without any change in technology and structural change. However, total CO2 emission in the year was only 385 million tonnes.. This indicates that the output increased without proportionate increase in CO2 emissions. This is due to structural shift or abatement measures or both. Total increase in CO2 emission is only 202 Mt between 1992 and 2005. It is 156 Mt less than the value of emission that would have resulted due to only activity level effects while all the other factors remain constant over time. This means that the trend in the manufacturing sector shows improvement in CO2 emission intensity with structural and technological changes. This decreasing trend in CO2 intensity is shown for all industries except aluminum, cement, machinery and textiles. Decomposition of total energy consumption and CO2 emissions Energy consumption Increase or decrease in energy consumption in general, is an indicator of energy efficiency. However, higher energy use does not always imply less efficient use of energy. Energy consumption trends are driven by changes in activity level (output level in the industry), change in energy efficiency (change in technologies known as intensity effect) and due to change in product mix (structural change). The output effects and impact of structure should be isolated to determine ‘pure’ energy intensity changes. In the present study, we applied total differentiation decomposition analysis. Decomposition methods attempt to separate all the structural effects, intensity effects and activity level effects. This approach is effective at separating and identifying the relative contributions of different effects. The results of energy consumption decomposition analysis are presented in Table 11. The changes in 8 Industries which use efficient technologies, fuel switch and good energy management practice. 9 This refers to change in product mix in the economy which will change the shares of different industries.

energy consumption shares are: 165.5%, −19.6% and −45.9% which originate in output, intensity and structural effect, respectively. The structural effect measures the changes in product mix in the economy, which was induced by changes in the composition of manufacturing sub-sectors. Economies producing large amounts of energy-intensive products like iron and steel, non-ferrous metals and cement are expected to consume more energy per unit of output than those with structures favoring less energy-intensive industries like electronics and textile industries. In general, if the structural effect is positive, there is a movement towards more energyintensive industries and energy intensity increases compared with the base year. The pure intensity effect measures improvements in energy efficiency, changes in technology, fuel mix changes, efficient energy management practice as well as any other factor which is not related to volume of output or composition. If this effect is positive, then it implies a worsening energy efficiency scenario. A negative pure intensity effect points to improvements in energy use. The output effect is positive when more output (in monetary terms) can be produced with the same energy use. So, if this effect is positive, then it implies an improvement in energy efficiency. Fig. 1 shows the annual trends in pure energy consumption at the macroeconomic level in chaining and decomposition between two consecutive years. The analysis shows that energy consumption is influenced mostly by structural change and level output change. Structural effect (SE) has negative impact and output effect (OE) positive impact on energy consumption. The negative impact of pure energy intensity effect (IE) is also shown in the figure. This means that OE is pulled back by the SE and IE. Hence, the overall effect on energy consumption does not improve to the extent that it would have with only OE. Industry-wise, IE and SE have mostly shown negative impacts on energy consumption offsetting the positive impacts of OE (see Table 11). Consequently, the real energy consumption increase is only 1422 PJ. The intensity effect in all the industries, except the textile and mining sectors, is negative. The structural effect also has negative impact on total energy consumption, except for chemicals, copper, food products and other nonmetals. This means that for the period of the study, 1992–2005, the overall structural and intensity effects are negative. This is due to a shift in output shares, from highto low energy-intensive industries and overall gain in energy efficiency. Energy intensity decline of 27.3, 21.2 and 4.8%, were realized in iron and steel, copper, and cement, respectively. The structural effect has more impact than the intensity effect (approximately twice) on changes in total energy consumption. During 1992–2005, there was a significant improvement in energy efficiency among iron and steel, chemical, aluminum, other metals, cement and paper industries. The textile industry has significant

B.S. Reddy, B.K. Ray / Energy for Sustainable Development 14 (2010) 35–47

43

Fig. 1. Decomposition of change in total energy consumption (1992–2005).Note: OE = output effect, IE = intensity effect, and SE = structural effect.

(positive) effect on increase in energy consumption by output effect and intensity effect but the structural effect is negative. Overall, at the aggregate level, the intensity effect is negative because the sum of decreasing intensity effect due to iron and steel, cement, chemical, aluminum and nonmetals has more than the increasing effect of textile and mining industries. Carbon dioxide emissions In the case of CO2 emissions, keeping the ‘pure’ intensity and structural effects fixed at the base year, the quantity of CO2 emissions

would be 541 MT to maintain the increased level of output. Fig. 2 and Table 12 show that output effect increases the CO2 emissions while the structural effect and pure intensity effect have reverse effect. For some years, pure intensity effect has positive influence, but the negative effect dominates. Industry-wise, the CO2 intensity effect in all the industries, except aluminum, beverages and tobacco, copper, machinery, other nonmetal and textiles is negative. This is attributed to overall gain in energy efficiency in those industries. Also, the structural effect is negative due to a shift in output shares, from high- to low energy-intensive industries. The changes in CO2 emission share at aggregate level are: 152,

Fig. 2. Decomposition of changes in total emissions (MTCO2) during 1992–2005.

44

B.S. Reddy, B.K. Ray / Energy for Sustainable Development 14 (2010) 35–47

Table 12 Decomposition of change in CO2 emission sector wise ′MTCO2 (1992–2005). Industry

Aluminum Beverage and tobacco Cement Chemicals Copper Food products Iron and steel Machinery Mining Other nonmetals Paper Textiles Transport equipment Total

Actual value

% Shares in total industry

% Shares in total effect

OE

IE

SE

Total

OE

IE

SE

Total

OE

IE

SE

37.7 0.5 54.0 49.0 0.7 8.1 91.5 0.5 0.5 12.3 22.0 30.0 0.6 307.3

1.0 0.1 0.3 − 13.1 0.0 − 2.8 − 24.1 0.0 0.7 − 18.2 − 2.7 33.1 − 0.1 − 25.9

− 15.7 − 0.1 − 26.4 6.1 0.1 3.4 − 43.6 0.0 0.0 5.4 − 1.0 − 7.5 0.0 − 79.5

23.0 0.4 27.8 42.0 0.8 8.6 23.8 0.5 1.2 − 0.5 18.3 55.6 0.4 201.9

12.3 0.2 17.6 15.9 0.2 2.6 29.8 0.2 0.2 4.0 7.2 9.8 0.2 100.0

− 3.8 − 0.3 − 1.0 50.5 0.0 10.9 93.1 − 0.1 − 2.8 70.4 10.5 − 128.0 0.5 100.0

19.7 0.1 33.2 − 7.6 − 0.1 − 4.2 54.8 0.0 0.0 − 6.8 1.3 9.4 0.0 100.0

11.4 0.2 13.8 20.8 0.4 4.3 11.8 0.3 0.6 − 0.3 9.1 27.5 0.2 100.0

163.8 106.4 194.2 116.7 88.6 93.9 384.3 92.1 41.7 − 2290.8 120.4 53.9 138.0 152.2

4.2 16.9 0.9 − 31.1 0.5 − 32.8 − 101.2 3.7 61.0 3393.0 − 14.8 59.6 − 30.2 − 12.8

− 68.1 − 23.3 − 95.1 14.4 10.9 38.9 − 183.1 4.2 − 2.7 − 1002.2 − 5.5 − 13.5 − 7.8 − 39.4

Note. OE = output effect; IE = intensity effect; and SE = structural effect.

− 13 and − 39% for the output, intensity and structural effects, respectively. Energy intensity The trend of energy intensities over the study period 1992–2005 is presented in Table 13 and Fig. 3. In the analysis of intensities the output is already controlled, and the change in intensity is a combination of IE and SE. Fig. 3 shows the energy intensity decomposition analysis between consecutive years for the period 1992–2005. The energy intensity decreased during the study period except for 1997–1998. The SE makes a positive impact on reducing the overall energy intensity except for the period 1995–1996. For the entire study period, SE has higher impact in reducing the overall energy intensity than the IE effect. The figure clearly shows that IE has a negative effect during the period 1992–1997 which turns positive during 1997–2005. The overall intensity fell from 804 (TJ/Rs billion) to 540 (PJ/Rs billion) between 1992 and 2005. This decrease in energy intensities is due to the contribution of the structural and intensity effects. The contribution of intensity and structural effects in this decrease are: −79.4 (30% of total) and −184.8 (70% of total), respectively. This means that in the overall decrease in energy intensity, a significant contribution emanates from structural effect. It is important to identify the exact factor responsible for energy intensity variation for each industry. Industry-wise, decreasing trends are noted for iron and steel, aluminum, cement, and tobacco and beverages. The largest reductions

in energy intensity have been achieved in iron and steel, cement, aluminum and paper industry categories. The pure intensity effect in all the industries, except for textile and mining sectors, is negative. This means that the negative intensity effect dominates during the period 1992–1997 and positive intensity effect during 1997–2005. The structural effect has negative impact on total energy consumption during both periods. Fig. 4 shows the variation of economic energy intensity for different fuels over the study period. This implies that the inter-fuel substitution has taken place which causes variation in total energy consumption and energy intensity. As we have seen, overall energy economic intensity falls over the years implying effect of substitution having a positive impact on energy conservation, energy efficiency and CO2 emission intensity. However, for some carriers energy intensity increases, but overall decrease is the order. Solid fuels and electricity energy intensities decrease by 55 and 22%, respectively, whereas those of gas and petroleum products increase by 109 and 37.5%, respectively (Fig. 5). Carbon dioxide intensity The industrial sector expanded during the study period, but the intensity of carbon dioxide emission decreased from 102 to 72.5 ′000 tCO2/Rs billion. The contribution of each factor to carbon dioxide intensity is shown in Table 14 and Fig. 6. Fig. 6 shows year-wise overall change in emission intensity with emission intensity effect and structural effect. Total combined effect is always negative except for

Table 13 Decomposition of energy intensity in TJ/Rs Billion (1992–2005). Industry

Aluminum Beverage and tobacco Cement Chemicals Copper Food products Iron and steel Machinery Mining Other Nonmetals Paper Textiles Transport Equipment Total

Actual unit

Shares in total industry (%)

Shares in total effect (%)

IE

SE

Total

IE

SE

Total

IE

SE

− 2.2 − 0.1 − 5.9 − 29.8 − 0.6 − 1.9 − 68.6 − 0.1 0.8 − 16.6 − 2.0 48.2 − 0.5 − 79.4

− 34.5 − 0.2 − 58.1 15.2 0.3 4.1 − 105.7 0.0 0.0 6.9 − 2.2 − 10.6 − 0.1 − 184.8

− 36.6 − 0.2 − 64.0 − 14.6 − 0.3 2.1 − 174.3 − 0.1 0.7 − 9.7 − 4.2 37.5 − 0.6 − 264.2

2.7 0.1 7.5 37.5 0.8 2.4 86.4 0.2 − 1.0 20.9 2.6 − 60.7 0.7 100.0

18.6 0.1 31.4 − 8.2 − 0.2 − 2.2 57.2 0.0 0.0 − 3.7 1.2 5.7 0.0 100.0

13.9 0.1 24.2 5.5 0.1 − 0.8 66.0 0.0 − 0.3 3.7 1.6 − 14.2 0.2 100.0

5.9 21.1 9.3 204.5 206.1 − 90.9 39.4 133.7 105.5 170.9 48.5 128.3 90.1 30.0

94.1 78.9 90.7 − 104.5 − 106.1 190.9 60.6 − 33.7 − 5.5 − 70.9 51.5 − 28.3 9.9 70.0

Note. OE = overall effect; IE = intensity effect; and SE = structural effect.

B.S. Reddy, B.K. Ray / Energy for Sustainable Development 14 (2010) 35–47

45

Fig. 3. Decomposition of changes in energy intensity in TJ/Rs billion (1992–2005) (base price 1991–1992).

the period 1997–1998. The contribution of intensity and structural effects in this decrease are: −7.1 (24.3% of total) and −22.2 (75.7% of total), respectively. This means that in the overall decrease in energy intensity, a significant contribution comes from structural effect. Approximately 23% of inter-fuel substitution is responsible for decrease in energy and CO2 intensity during the entire period (1992–2005). That is reflective of industry shifting to high efficiency fuel from a low efficiency one. Policy implications The quest to catch on the lagged economic growth and rising population in developing countries has propelled the demand for energy. The rise in energy consumption brings to light two crucial national issues, the issue of foreign exchange and environment. An indicator of energy use is energy intensity calculated at the macro-

economic level using aggregated data from micro level. It shows how various effects, structural, technological or output, help change aggregate energy use over a certain period of time. However, this indicator provides general information on energy consumption trends and provides insight into energy efficiency. In recent times, the policy focus has shifted to environmental implications associated with energy use. The debate on greenhouse gas emissions and their role in stimulating global climate change focuses on the efficient use of energy in various countries. The amount of fossil fuel burnt energy consumed by a country and the efficiency of its usage are two major factors determining a country's overall CO2 emissions. In other words, policy-makers are becoming increasingly concerned about the environmental changes relative to economic repercussions of energy use. Economic intensity indicators provided in this study are critical in terms of information they can provide to policy-makers. However, measuring the true effectiveness of a policy requires careful monitor-

Fig. 4. Inter-fuel substitution effect on energy intensity (1992–2005).

46

B.S. Reddy, B.K. Ray / Energy for Sustainable Development 14 (2010) 35–47

Fig. 5. Inter-fuel substitution effect on CO2 intensity (1992–2005).

ing and comparison to pre-policy data. Complete decomposition technique has been used here to help and evaluate the relative contribution of factors that effect changes in total energy consumption, energy intensity, carbon dioxide emission and carbon dioxide intensity. It helps in identifying the main components of change in energy use and carbon dioxide emission. This provides greater insights into energy efficiency levels in various sectors of the economy to appreciate the way the structural and other effects influence energy use. The method produces indicators of “pure” energy intensity by separating out the effects of structural change. The method is relatively easy to apply, and typically needs data that is readily available. In general, the factors behind the decrease in energy intensity are structural and technology effects, with structural effect being more than technological. The increase in total energy consumption is mainly due to output effect which is always positive. A similar pattern has been observed in CO2 emissions. The study has shown significant influence of structural change which implies that technology alone is not responsible for comparative reduction in energy consumption and carbon dioxide emission. The movement from a higher energyintensive sub-sector to a lower one would matter in the energy economy. Hence, inter-industrial macro policies are more important than intra-industrial micro policies. Conclusions Energy intensity indicators help design energy policies which can be used in the analyses of efficiency improvements. It is recom-

mended that such indicators be used to get a broad idea of how energy use is changing in the economy. Decisions regarding the choice of an appropriate energy efficiency strategy, either market-based or command control, be evaluated using such indicators. The trends in carbon dioxide intensity and the major factors that affect it (structural and economic changes) provide policy-makers with the information needed to set CO2 targets for various industries; as also design appropriate CO2 abatement strategies. Hence, these serve not just as monitoring tools, but also as a basis for energy efficiency policies and regulations aimed at achieving greater energy conservation and pollution abatement measures. This energy or emission intensity study at individual industry level is more useful than in overall manufacturing energy/carbon intensity study. This study helps in identifying the industry where potential exists for efficiency improvement and pollution abatement. Unlike most of the sub-sectors, the textile industry increased energy use and CO2 in the study period because of mechanization. However, as technology matures in future, the intensity is expected to decrease. In the paper industry, there is a decline in the intensity since 2000 and it is evident that improvement is ongoing. Decomposition of the energy intensity effects observed in the fuel substitution in the sector indicated that the structural effect (SE) was the main contributor to the changes in energy intensity. It has 2.3 times higher influence in reducing energy consumption and 3.1 times in reducing CO2 emission than intensity effect (IE). This shows that the manufacturing sector has moved to more energy-efficient sub-sectors.

Table 14 Decomposition for Change in CO2 intensity (′000 tCO2/Rs billion) (1992–2005). Industry

Aluminum Beverage and tobacco Cement Chemicals Copper Food products Iron and steel Machinery Mining Other nonmetals Paper Textiles Transport equipment Total

Actual unit

Shares in total industry (%)

Shares in total effect (%)

IE

SE

Total

IE

SE

Total

IE

SE

0.3 0.0 0.1 − 3.6 0.0 − 0.8 − 7.1 0.0 0.2 − 4.9 − 0.8 9.5 0.0 − 7.1

− 4.4 0.0 − 7.4 1.7 0.0 1.0 − 12.6 0.0 0.0 1.7 − 0.3 − 1.9 0.0 − 22.2

− 4.1 0.0 − 7.4 − 1.9 0.0 0.2 − 19.6 0.0 0.2 − 3.2 − 1.1 7.6 0.0 − 29.4

− 4.0 − 0.3 − 1.0 51.0 0.0 10.8 99.1 − 0.1 − 2.8 69.3 10.7 − 133.2 0.5 100.0

19.8 0.1 33.5 − 7.8 − 0.1 − 4.4 56.5 0.0 0.0 − 7.7 1.3 8.7 0.0 100.0

14.0 0.0 25.1 6.5 − 0.1 − 0.7 66.8 0.0 − 0.7 11.0 3.6 − 25.7 0.1 100.0

− 6.9 − 295.5 − 1.0 191.9 4.3 − 383.5 36.0 47.2 104.1 152.6 72.6 125.7 79.2 24.3

106.9 395.5 101.0 − 91.9 95.7 483.5 64.0 52.8 − 4.1 − 52.6 27.4 − 25.7 20.8 75.7

B.S. Reddy, B.K. Ray / Energy for Sustainable Development 14 (2010) 35–47

47

Fig. 6. Decomposition of changes in emission intensity (1992–2005).

Appendix 1 Dollar–Rupees conversion factor for the analysis years. Calendar year

Rs per dollar

Dollar per Rs

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005

25.9206 31.4439 31.3742 32.4198 35.4280 36.3195 41.2665 43.0552 44.9401 47.1857 48.5993 46.5818 45.3165 44.1000

0.038579354 0.031802671 0.031873323 0.030845348 0.028226262 0.027533419 0.024232731 0.023225998 0.022251842 0.021192861 0.020576428 0.021467612 0.022067018 0.022675737

Source: Reserve Bank of India (2008). Appendix 2 Conversion factors. Fuel

tCO2/TJ

Solid fuels Petroleum products Gas Electricity

111 77 69 316

Note: the conversion factors for electricity is obtained from Das and Kandpal (1997) and other fuels are obtained from ETC/ACC (2003) and Earth Trends (2003). References Ang BW, Lee SY. Decomposition of industrial energy consumption: some methodological and application issues. Energy Econ 1994;16:83–92.

Bhattacharya RN, Paul S. Sectoral changes in consumption and intensity energy in India. Indian Economic Review 2001;XXXVI(2):381–92. Boyd GA, Hanson DA, Sterner T. Decomposition of changes in energy intensity: a comparison of the Divisia index and other methods. Energy Econ 1988;10:312. Choi Ki-Hong, Ang BW. Decomposition of aggregate energy intensity changes in two measures: ratio and difference. Energy Econ 2003;25:615–24. CMIE, 2007. Reports of Centre for Monitoring Indian Economy, Mumbai. Das A, Kandpal TC. Energy-environment implications of cement manufacturing in India: a scenario analysis. Int J Energy Res 1997;21:299–308. De Bruyn SM. Economic growth and the environment: an empirical analysis. Norwell: Kluwer; 2000. Earth trends, 2003, online data base. Available at earthtrends.wri.org. Edwards R, Parikh A. Intensities of energy usage an international and inter-temporal comparison. Energy Policy 1978;6:66–75. ETC/ACC, 2003. Comparison of CO2 emission factors for fuels used in Greenhouse Gas Inventories and consequences for monitoring and reporting under the EC emissions trading scheme Technical Paper 2003/10, The European Topic Centre on Air and Climate Change (ETC/ACC). Farla J, Blok K, Schipper L. Energy efficiency developments in the pulp and paper industry. Energy Policy 1997;25:745–58. GoI, 2008. Industrial and product classifications in India, Ministry of Statistics and Programme Implementation, Govt. of India. IEA. World energy outlook 2007. Paris: International Energy Agency; 2007. Li Jing-wen, Shrestha RM, Foell Wesley K. Structural change and energy use. Energy Econ 1990:109–15. Lin SJ, Chang TC. Decomposition of SO2, NOx and CO2 emissions from energy use of major economic sectors in Taiwan. The Energy Journal 1996;17:1-17. Nooji M, Kruk R, Soest DP. International comparisons of domestic energy consumption. Energy Econ 2003;25:259–373. Park Se-Hark. Decomposition of industrial energy consumption. Energy Econ 1992;13: 265–70. Reddy M. Energy consumption and economic activity in Fiji. J Pac Stud 1998;22: 81–96. Reddy BS. Economic and social dimensions of household energy use: a case study of India. In: Ortega E, Ulgiati S, editors. Proceedings of IV Biennial International Workshop “ Advances in Energy Studies”. Campinas, SP, Brazil: Unicamp; 2004. Reitler W, Rudolph M, Schaefer H. Analysis of the factors influencing energy consumption in industry: a revised method. Energy Econ 1987;9:145–8. Reserve bank of India (RBI), 2008. Handbook of Statistics on Indian Economy, RBI publication. Rose A, Casler S. Input–output structural decomposition analysis: a critical appraisal. Economic System Res 1996;8:33–62. Sun JW. Changes in energy consumption and energy intensity: a complete decomposition model. Energy Econ 1998;22:85-100.