Energy 68 (2014) 688e697

Contents lists available at ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

Analysis of energy-related CO2 (carbon dioxide) emissions

and reduction potential in the Chinese non-metallic mineral
products industry
Boqiang Lin a, b, *, Xiaoling Ouyang c
a
New Huadu Business School, Minjiang University, Fuzhou 350108, China
b
Collaborative Innovation Center for Energy Economics and Energy Policy, Institute for Studies in Energy Policy, Xiamen University,
Xiamen, Fujian 361005, China
c
School of Energy Research, College of Energy, Xiamen University, Xiamen 361005, China

a r t i c l e i n f o a b s t r a c t

Article history: As one of the six highest energy-consuming industries, non-metallic mineral products industry accounts
Received 23 October 2013 for 9.5% of national energy usage and 9.0% of energy-related CO2 (carbon dioxide) emissions in China. In
Received in revised form this paper, we evaluate the CO2 emissions change from energy consumption and present a compre-
11 January 2014
hensive picture of the underlying determinants of emissions change in the Chinese non-metallic mineral
Accepted 18 January 2014
Available online 20 February 2014
products industry for the period 1986e2010, based on the LMDI (Logarithmic Mean Divisia Index)
method. Results demonstrate that industrial activity is the leading force explaining emission increase
while energy intensity is the major contributor to the emission mitigation. Effects of industrial scale and
Keywords:
CO2 emissions change
carbon intensity of energy show varying trends interchanging intervals of growth along the study period.
Energy consumption Moreover, substitution effect has a small negative impact on the increase of CO2 emissions. Reduction
The LMDI method potential of CO2 emissions is predicted to be 99.02 Mt in 2020 under the scenario of moderate emission
Emission reduction potential reduction, which accounts for 0.30% of the world’s energy-related CO2 emissions in 2011; and 188.88 Mt
Non-metallic mineral products industry under the scenario of aggressive emission reduction, which accounts for 0.58% of the world’s total in
China 2011 and is close to energy-related CO2 emissions of Egypt in 2010. Policy recommendations are thus
made for future emission reductions.
Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction First, it is highly energy-intensive and is one of the highest energy-

consuming industries in China. Energy consumption of non-metallic
1.1. The research background mineral products industry accounted for 9.5% of national and
13.5% of industrial and 16.4% of manufactural energy usage during
Continual growth of CO2 (carbon dioxide) emissions in China has 1985e2011. Energy consumption of the Chinese non-metallic min-
been the center of attention both domestically and internationally eral products industry was 300.15 million tonnes of coal equivalent
[1e3]. The industrial sector is the largest primary energy-use sector (Mtce) in 2011, which was higher than the total energy consumption
in China that is extremely energy-intensive and responsible for of Iran (293.36 Mt) in the same year [5]. The huge amount of energy
associated CO2 emissions [4]. Considering the leading role that the consumption highlights its importance in the research area.
industrial sector has played in CO2 emissions in China, it is neces- Second, it is one of the most carbon-intensive industries in
sary to investigate the influencing factors of CO2 emissions change China. CO2 emissions from energy consumption in the non-metallic
from energy consumption in the major industrial subsectors so as to mineral products industry amounted to 544.82 Mt in 2011 e higher
provide recommendations to policy makers. than the energy-related CO2 emissions of the United Kingdom in
The non-metallic mineral products industry was chosen as the 2010 (528.89 Mt), accounting for 1.67% of the world’s and 6.3% of
study object for three reasons: China’s energy-related CO2 emissions [6]. Considering its pre-
dominance in China’s CO2 emissions, it is extremely important to
* Corresponding author. New Huadu Business School, Minjiang University, Fuz-
investigate the potential of CO2 emissions reduction.
hou, Fujian 350108, China. Tel.: þ86 5922186076; fax: þ86 5922186075. Third, it has a typical energy consumption structure. Fuel mix
E-mail addresses: bqlin@xmu.edu.cn, bqlin2004@vip.sina.com (B. Lin). of the Chinese non-metallic mineral products industry has been

http://dx.doi.org/10.1016/j.energy.2014.01.069
0360-5442/Ó 2014 Elsevier Ltd. All rights reserved.
 B. Lin, X. Ouyang / Energy 68 (2014) 688e697 689

dominated by coal, of which the consumption accounted for 7.3% of application and flexibility, Ang et al. [21] concluded that the LMDI
national, 7.7% of industrial and 19.5% of manufactural coal usage in methods, in particular LMDI I, have several advantages over other
2011. Considering the competitive price and relative abundant re- perfect decomposition methods in energy and environmental
sources of coal, coal-dominated energy structure is difficult to analysis. Aiming at handling zero values in the LMDI approach, Ang
reverse in the short term [7]; and it is clear that China’s future and Liu [22] compared the SV (small value) and AL (analytical limit)
economic growth is unsustainable [8]. Therefore, considering its strategies, and extended earlier works by generalizing the analyt-
representativeness, it is crucial to capture determinants of energy- ical limits of LMDI. In order to expand research, Ang et al. [23]
related CO2 emissions change in the Chinese non-metallic mineral analyzed the relationship between LMDI II and LMDI I, and
products industry. further supported the recommendation made by Ang [11] that
LMDI is the “preferred” IDA method.
1.2. The contribution of this study Applications of the LMDI method in investigating carbon
emissions have been initiated in a number of earlier studies. In this
Both energy consumption and the related CO2 emissions in the article, we review the recent studies as well as major determinants
Chinese non-metallic mineral products industry have shown up- of CO2 emissions change in chronological order. By applying the
ward trends over the years. On the other hand, China’s urbanization LMDI method, Wang et al. [24] analyzed the change of aggregated
and industrialization process will not come to an end until 2020 [9], CO2 emissions in China from 1957 to 2000. Results indicated that
implying that the industrial growth will still be the leading force energy intensity, fuel switching and renewable energy penetration
driving economic growth and energy consumption. The challenge exhibited positive effects to the decrease in CO2. Liu et al. [25]
of CO2 emissions reduction from the non-metallic mineral products analyzed the change of industrial carbon emissions from 36 in-
industry is particularly important because of its key role in ur- dustrial sectors in China over the period 1998e2005 and showed
banization process and the realization of China’s transition into a that the overwhelming contributors to the change of China’s in-
low-carbon economy. In this context, this paper intends to address dustrial sectors’ carbon emissions were industrial activity and en-
the following important questions: First, what are the major factors ergy intensity. Ma and Stern [26] focused on the fall and resurgence
that influence CO2 emissions change from energy consumption in in China’s carbon emissions since the mid-1990s. Results demon-
the Chinese non-metallic mineral products industry? Second, what strated that the positive effect of population growth had been
degree of influence of each factor? Third, how much emission decreasing over the period 1971e2003. Malla [27] examined the
reduction potential does the non-metallic mineral products in- role of three factors (electricity production, electricity generation
dustry have? Fourth, what policy implications can we draw in the structure and energy intensity of electricity generation) affecting
aspect of mitigating CO2 emissions from industrial sectors in China? the evolution of CO2 emissions from electricity generation in seven
Answers to these questions are meaningful for the low-carbon AsiaePacific and North American countries. Results showed that
transition strategy in China as well as the sustainable develop- production was the major factor responsible for rise in CO2 emis-
ment of industrial sectors. China’s economic and industrial devel- sions while energy intensity was responsible for modest reduction
opment will face the issue of transition when economic growth in CO2 emissions during the period 1990e2005. Zhao et al. [28]
slows down after 2020. The industrial development will no longer decomposed the influencing factors of ICE (industrial carbon
be extensive and supported by wasteful energy consumption; more emissions) in Shanghai, China. Results demonstrated that the in-
importantly, it will be constrained by a low-carbon economy. dustrial output was the main driving force of ICE while energy
The remainder of this paper is structured as follows. Section 2 intensity, energy adjustment and industrial structure were major
presents the literature review. Section 3 describes the methodol- determinants for reduction of ICE. Zha et al. [29] investigated fac-
ogy and data source and provides an overview of the industry. tors that may affect the changes of energy-related CO2 emissions
Section 4 discusses the empirical results of our study. Section 5 from urban and rural residential energy consumption during 1991e
summarizes our findings and provides policy recommendations. 2004. Results showed that energy intensity and the income effects,
respectively, contributed most to the decline and the increase of
2. Literature residential CO2 emissions for both urban and rural China. Oh et al.
[30] analyzed the specific trends and influencing factors that had
With the global warming becoming a serious issue in the world, caused changes in emission patterns in economic sectors in South
decomposition analysis of CO2 emissions has been a hot issue in Korea over 1990e2005. Results indicated that economic growth
academia since the 1980s. In the application areas, there have been was a dominant explanation for the increase in CO2 emissions in
more studies dealing with industrial energy usage and energy- manufacturing, services and residential sectors, while energy in-
related gas emissions [10]. The advantages of the IDA (index tensity was an important driver of CO2 reduction in most sectors
decomposition analysis) are that it assesses the driving forces that except for several manufacturing sub-sectors. Akbostancı et al. [31]
underlie CO2 emissions change and presents implications for CO2 decomposed the changes in the CO2 emissions of Turkish
emissions mitigation [11]. There are a set of decomposition manufacturing industry and found that industrial activity and en-
methods linked to the Divisia index [12,13] and the Laspeyres index ergy intensity were the primary factors determining CO2 emissions
[14] in the index decomposition analysis. Within the Divisia index, change. Zhang et al. [32] decomposed energy-related carbon
the LMDI (logarithmic mean Divisia index) method was first emissions of 30 provinces in China during 1995e2009 by four
introduced by Ang and Choi [15], Ang et al. [16], which can be causal factors e GDP (gross domestic products), economic struc-
conveniently applied in comparative studies of cross-country/ ture, energy intensity, and fuel mix. Results demonstrated that GDP
region energy and environmental issues [17,18]; therefore, it was and energy intensity were the major factors contributing to the CO2
recommended as the preferred method because of its theoretical emissions change. Liu et al. [33] explored driving forces of China’s
foundation, adaptability, ease of use and result interpretation [11]. regional and sectoral GHG emission. Results indicated that the
Ang [19] further presented a useful guide of the LMDI approach. In inequity of technology level among regions was a main barrier for
2001, Ang and Liu [20] presented a new energy decomposition China’s CO2 mitigation. Hammond and Norman [34] used LMDI
method, called the Log-Mean Divisia Index Method I (LMDI I) and method to separate the contributions of changes in output, indus-
concluded that the method has the desirable properties of perfect trial structure, energy intensity, fuel mix and electricity emission
decomposition and consistency in aggregation. In terms of ease of factor to the reduction in energy-related carbon emissions from UK
690 B. Lin, X. Ouyang / Energy 68 (2014) 688e697

manufacturing. Results showed that the reduction in energy in- 3.1.2. Decomposition method
tensity was the primary reason for the fall in emissions between This paper adopts the Logarithmic Mean Divisia Index (LMDI)
1990 and 2007. Fujii et al. [35] analyzed the effects of the changes method to analyze CO2 emissions change from energy consump-
made in China’s industrial sector on the three main air pollutants tion. The change in CO2 emissions of the Chinese non-metallic
between 1998 and 2009. Shao et al. [36] analyzed carbon emissions mineral products industry (DCO2) between a base year 0 and an
from energy consumption of the industrial sector in City Tianjin, end year T can be decomposed into the effects of (i) the change in CI
China. Results showed that improvements in energy utilization (carbon emissions per unit of fossil-fuel, termed the carbon in-
efficiency were the most important contributors to effective in- tensity effect, CIeff); (ii) the change in S (the share of fossil fuels in
dustrial energy conservation and emission reductions. total energy, termed the substitution effect, Seff); (iii) the change in
From the present literature, although the application of LMDI EI (energy consumption per unit of industrial value added, termed
method is mainly in the area of industrial energy and environmental the energy intensity effect, EIeff); (vi) the change in IA (the indus-
analysis, research on energy-related CO2 emissions in China’s in- trial value added per worker, termed the industrial activity effect,
dustrial sectors is still limited. Moreover, to the best of our knowl- IAeff) and (v) the change in IS (employment; termed the industrial
edge, no studies have been conducted on CO2 emissions change in structural effect, ISeff):
the Chinese non-metallic mineral products industry in spite of its
important role in China’s energy demand and related CO2 emissions. DCO2 ¼ CO2 ðTÞ  CO2 ð0Þ ¼ CIeff þ Seff þ EIeff þ IAeff þ ISeff (4)
The effects, in turn, can be calculated from the following formula
3. Methodology and data source using the LMDI method:

3.1. Decomposition analysis CO2 ðTÞ  CO2 ð0Þ

CIeff ¼ *ln½CIðTÞ=CIð0Þ (5)
ln½CO2 ðTÞ=CO2 ð0Þ
3.1.1. Model construction
The following symbols are defined, in order to present the CO2 ðTÞ  CO2 ð0Þ
methodology used in this paper: Seff ¼ *ln½SðTÞ=Sð0Þ (6)
ln½CO2 ðTÞ=CO2 ð0Þ

TOE ¼ total energy consumption

CO2 ðTÞ  CO2 ð0Þ
GDP ¼ the gross domestic product EIeff ¼ *ln½EIðTÞ=EIð0Þ (7)
POP ¼ population ln½CO2 ðTÞ=CO2 ð0Þ
EFF ¼ energy consumption from fossil fuels
CI ¼ CO2 emissions per unit of fossil-fuel CO2 ðTÞ  CO2 ð0Þ
IAeff ¼ *ln½IAðTÞ=IAð0Þ (8)
S ¼ the share of fossil fuels in total energy ln½CO2 ðTÞ=CO2 ð0Þ
EI ¼ energy consumption per unit of IVA (industrial value
added) CO2 ðTÞ  CO2 ð0Þ
IA ¼ industrial value added (IVA) per worker ISeff ¼ *ln½ISðTÞ=ISð0Þ (9)
ln½CO2 ðTÞ=CO2 ð0Þ
IS ¼ employment1
Based on the above equations, CO2 emissions change from energy
The decomposition of fossil fuel CO2 emissions into related consumption in the Chinese non-metallic mineral products industry
factors dates back to a series of studies undertaken in the 1980s, can be calculated. We use relevant data over 1986e2010 and sepa-
mainly at industry level for a single industrialized country [37]. rate the period into five time intervals for easier data management:
Kaya [38] proposed the Kaya Identity and decomposed the CO2 1986e1990, 1991e1995, 1996e2000, 2001e2005 and 2006e2010
emissions into several affecting variables: [7]. The set of intervals is reasonable because time intervals of the
national economic development plans in China are five years [32].
CO2 TOE GDP In order to estimate the reduction potential of energy-related
CO2 ¼ * * *POP (1)
TOE GDP POP CO2 emissions, we further construct the decomposition model.
According to equation (4), if the effects of CO2 emissions per unit of
Equation (1) established the relationship between carbon di-
fossil-fuel (CIeff), substitution (Seff), energy intensity (EIeff), indus-
oxide emissions and determinants of CO2 emissions change such as
trial activity (EIeff) and industrial scale (ISeff) from a base year 0 to a
carbon dioxide emissions per unit of energy usage, energy utiliza-
target year T can be estimated, the CO2 emissions in year T therefore
tion efficiency, level of economic development and population. In
can be forecast as equation (10):
this article, equation (1) has subsequently been expanded as:
CO2 ðTÞ ¼ CO2 ð0Þ þ CIeff þ Seff þ EIeff þ IAeff þ ISeff (10)
COt2 EFFt TOEt GDPt
COt2 ¼ * * * *POPt (2)
EFFt TOEt GDPt POPt We assume that a, b, d, f and g are the change rates of the CI
(carbon intensity) effect, S effect, EI effect, IA effect and IS effect
Identities in equation (2) focus on CO2 emissions from the
from a base year 0 to a target year T, respectively. Therefore,
combustion of fossil fuels. For a specific industrial sector, the above
CI(T) ¼ CI(0)*(1 þ a), S(T) ¼ S(0)*(1 þ b), EI(T) ¼ EI(0)*(1 þ d),
equation can be rewritten as:
IA(T) ¼ IA(0)*(1 þ f), IS(T) ¼ IS(0)*(1 þ g). Thus,
COt2 EFFt TOEt IVAt t CIeff ¼ W*lnð1 þ aÞ (11)
COt2 ¼ * * * *IS ¼ CI*S*EI*IA*IS (3)
EFFt TOEt IVAt ISt
Seff ¼ W*nð1 þ bÞ (12)

EIeff ¼ W*lnð1 þ dÞ (13)

1
Generally, an increase in employment implies the expansion of an economic
sector, so that the indicator of employment in this paper is used to represent the
industrial scale effect. IAeff ¼ W*lnð1 þ 4Þ (14)
 B. Lin, X. Ouyang / Energy 68 (2014) 688e697 691

ISeff ¼ W*lnð1 þ gÞ (15)

where

CO2 ð0Þ*½ð1 þ aÞ*ð1 þ bÞ*ð1 þ dÞ*ð1 þ 4Þ*ð1 þ gÞ  1

W ¼ (16)
ln½ð1 þ aÞ*ð1 þ bÞ*ð1 þ dÞ*ð1 þ 4Þ*ð1 þ gÞ
Based on the above model, future CO2 emissions can be pre-
dicted and contributions of each factor can be quantified.

3.2. Data source and industry overview

3.2.1. Data source

This article is based on annual data covering the period from
1986 to 2010. All data are collected from China Statistical Yearbook
Fig. 1. The trends of fossil-fuel energy consumption and CO2 emissions in the Chinese
[39]. non-metallic mineral products industry during 1985e2010.
Carbon dioxide emissions from fossil fuel consumption are Source: China Energy Statistical Yearbook [40], CEIC China Database [42].
calculated by:

X
16 X
16 Productions of other major products such as plate glass and ce-
CO2 ¼ CO2 ¼ Ei *CFi *CCi *COFi *ð44=12Þ (17) ramics also ranked top in the world.
i¼1 i¼1 Production growth in the non-metallic mineral products in-
In which, CO2 stands for the energy-related carbon dioxide dustry has led to rapid increase in energy consumption. During the
emissions; i represents the i-th energy; CFi is the conversion factor period 1985e2010, fossil fuel energy consumption (excluding elec-
from physical unit to kJoule (in this paper, the calculation took the tricity) increased from 77.47 Mtce to 244.63 Mtce; energy-related
Appendix IV in China energy statistical yearbooks for reference carbon dioxide emissions grew from 185.05 Mt to 507.29 Mt, an
[40]); CCi is carbon content coefficient of fuel type i, which is increase of 174.14%. The trends of fossil-fuel energy consumption
collected from the Intergovernmental Panel on Climate Change and CO2 emissions in the Chinese non-metallic mineral products
[41]; COFi is the carbon oxidation factor; and 44/12 is the conver- industry during 1985e2010 are shown in Fig. 1.
sion factor from carbon to carbon dioxide. In summary, CFi*CCi*- The average annual growth rate of fossil-fuel energy consump-
COFi*44/12 is the coefficient of carbon dioxide emissions for the i-th tion was 4.98% during 1985e2010, and energy-related CO2 emis-
energy, which is assumed constant over time. In order to avoid the sions were drove up by 4.49% annually. It is worth noting that both
double counting of CO2 emissions from the power generation in- fossil fuel energy consumption and the related CO2 emissions
dustry, electricity consumption is regarded as clean. increased significantly after 2002 due to China’s accelerated pro-
However, we should note the limitations of the LMDI method- cess of urbanization. The average annual growth rate of fossil-fuel
ology in the aspect of application due to the use of current value for energy consumption during 2003e2008 reached 12.41%, which
energy content, conversion rates of different energy sources from drove up the growth of related CO2 emissions at the rate of 12.08%.
physical unit to kJoule, and emission factors in defining the emis- Energy consumption structure of the Chinese non-metallic
sion performance of energy in a time-series analysis [7]. mineral products industry was dominated by coal, of which the
consumption increased from 86.14 Mt in 1985 to 250.32 Mt in 2010
with an annual growth rate of 4.63%. The growth rates of natural
3.2.2. Industry overview gas and electricity consumption were the highest among different
Non-metallic mineral products industry is a pillar industry in types of energy. Specifically, natural gas consumption increased
China, of which the value added contributes to nearly one percent from 180 million cubic meters in 1985e6376.4 million cubic meters
of the GDP (gross domestic products) each year [42]. In 2011, main in 2011 with the annual growth rate of 24.60%; electricity con-
business income of the non-metallic mineral products industry sumption grew from 22.16 billion kWh to 291.79 billion kWh dur-
amounted to 3929.48 billion CNY, accounting for 4.67% that of the ing the same period, of which the annual growth rate was 10.58%.
industrial sector [43]. Particularly, it greatly supports the under- The significant decline in coal consumption during 2000e2002 was
going rapid urbanization process in China, which propels the large- mainly due to the industrial restructuring, during which, a large
scale construction of urban infrastructure and housing and requires number of small companies went bankrupt. However, coal con-
huge quantities of cement. As the world’s largest cement consumer, sumption rose sharply back after 2003 due to the fixed asset in-
China consumed 2048 metric tons of cement in the year 2011. The vestment growth and the overheated real estate industry. As shown
large quantity of cement could only have been produced at home, in Fig. 2, coal is the major contributor to CO2 emissions, followed by
because no other country could afford supplying such large-scale fuel oil and coke. During 1985e2011, the average proportion of CO2
raw materials [44]. On the other hand, the growth of the Chinese emissions from coal consumption in the total CO2 emissions in the
non-metallic mineral products industry is driven by the infra- Chinese non-metallic mineral products industry was up to 90.60%;
structure boost and will be further pushed ahead by national policy and the average proportions of CO2 emissions from the consump-
[45]. The industrial valued added (IVA) increased greatly over the tion of fuel oil and coke were 3.49% and 2.52%, respectively.
past three decades from 20.79 billion CNY in 1985 to 337.76 billion It can be seen from Fig. 2, coal played a key role in the total CO2
CNY in 2010 (at 1990 constant prices). The average annual growth emissions in the Chinese non-metallic mineral products industry.
rate of IVA was 12.82% during 1985e2010, which was higher than Most of the increase in the CO2 emissions during the study period
that of the GDP. Take the production volume of cement for example. derived from the growth of coal consumption and coal-dominated
Since 1985, China has kept the record of the world’s largest cement energy structure. Similarly, the decrease in CO2 emissions during
producer. Cement output increased from 144.28 Mt in 1985e the period of 2000e2002 (industrial restructuring) was driven by the
2184.04 Mt in 2010 e equivalent to a growth of 1413.66% [43]. reduction of coal consumption. Energy diversification helped to
692 B. Lin, X. Ouyang / Energy 68 (2014) 688e697

during 1986e2010 (Fig. 3) for the following two reasons: first, China’s
national economic planning, which has significant impacts on the
macro-economy as well as the energy system, is based on the five-
year time interval; second, the impact of each factor on CO2 emis-
sions change among time intervals can be captured and compared.
Results show that there have been significant CO2 emissions
changes in the Chinese non-metallic mineral products industry
over the last 26 years. Identifying reasons of CO2 emissions change
would be helpful for policy design on mitigating the increase in CO2
emissions. As shown in Fig. 3, IA (industrial activity) effect was the
main force that led to the increase in CO2 emissions while EI (en-
ergy intensity) effect was the major contributor to the decline in
CO2 emissions. The IS (industrial scale) effect on CO2 change was
unstable among time intervals. In general, it contributed to the
increase in CO2 emissions. However, the IS effect led to the signif-
icant decline in CO2 emissions from 1996 to 2000 because of in-
Fig. 2. Carbon dioxide emissions from energy consumption in the Chinese non-
dustrial restructuring. During the restructuring period, a large
metallic mineral products industry. number of private and small enterprises as well as the state-owned
Source: China Energy Statistical Yearbook [40]. companies went bankrupt, and lots of workers were laid off.
Shrinking industrial scale resulted in the decline in energy con-
sumption as well as the corresponding CO2 emissions. Moreover,
reduce the share of coal. Natural gas consumption increased
the substitution (S) effect contributed to CO2 emissions decrease
dramatically after 2004, of which the average growth rate during
even though the degree of influence was quite small. The effect of
2004e2011 was as high as 67.57%. During the same period, the
carbon intensity of energy (CI) showed a varying trend inter-
growth of electricity consumption reached 13.68%. In conclusion, the
changing intervals (increasing and decreasing) during 1986e2010;
substitution of cleaner energy to coal helped to mitigate the growth of
however, it was the cause for the decrease in CO2 emissions. The
CO2 emissions in the Chinese non-metallic mineral products industry.
above main results are consistent with the study of Ang [46], which
showed that energy intensity (energy/GDP) had a greater impact on
4. Results and discussion energy-related carbon emissions than carbon intensity of energy
(carbon/energy).
4.1. Carbon dioxide emissions change during time intervals In order to conduct in-depth analysis, we analyze the influence of
each effect during the study period by degree. We observe that the
Decomposition analysis can quantify the impact of each deter- biggest contributor to the emissions increase was industrial activity
minant on CO2 emissions change from energy consumption by de- (IA) effect, which contributed 224.34 Mt CO2 during 1996e2000 and
gree. In this article, the study period is split into five time intervals 217.44 Mt CO2 during 2001e2005, the highest amounts during the

Fig. 3. Decomposition of CO2 emissions changes from energy consumption in the Chinese non-metallic mineral products industry.
 B. Lin, X. Ouyang / Energy 68 (2014) 688e697 693

time intervals. Even though the impact of IA showed a decreasing

trend after the time interval of 1996e2000, it was still the leading
force that pushed up the CO2 emissions in the Chinese non-metallic
mineral products industry. The energy intensity (EI) effect was the
major contributor to the decrease in emissions, of which the degree
of influence showed an increasing trend over the time intervals
from 17.54 Mt CO2 during 1996e2000 to 179.27 Mt CO2 during
2006e2010. The EI effect was quite significant during 2006e2010,
which was attributed to the national energy intensity reduction
target. In 2009, the Chinese government committed to reduce
energy intensity by 40%e45% in 2020 compared to the level in
2005. This target has been decomposed into three time intervals of
2009e2010, 2011e2015 and 2016e2020 for realization. Therefore,
we can expect that the EI effect would play a more important role in
reducing CO2 emissions in the Chinese non-metallic mineral prod- Fig. 4. Comparison of CO2 emissions and industrial value added trends in the Chinese
ucts industry in the future. The IS (industrial scale) effect, the second non-metallic mineral products industry.
largest influential factor of the CO2 emissions increase, contributed Source: China Statistical Yearbook [40].
117.68 Mt to CO2 emissions increase during 2006e2010.

11th Five-year Plan (2006e2010), which committed to reduce en-

4.2. An in-depth analysis of major influencing factors
ergy intensity by 20% in 2010 compared to the level in 2005. How-
ever, the energy intensity reduction target was not realized by the
Results of this paper demonstrated that the industrial activity
effective planning such as changing energy structure, improving
(IA) effect and the energy intensity (EI) effect were the leading
energy efficiency, promoting energy pricing reform, etc. As a matter
forces responsible for the CO2 emissions increase in the Chinese
of fact, it was achieved by power rationing on high energy-
non-metallic mineral products industry. An in-depth analysis is
consuming industries such cement and steel industries in the late
conducted for the purposes of investigating possible reasons of CO2
2010. Moreover, to a certain extent, the 2008 financial crisis also
emissions change, providing policy implications for mitigating CO2
helped to reduce energy intensity in industrial sectors. Therefore, the
emissions, and formulating development strategies for the Chinese
decreased CO2 emissions contributed by the EI effect during 2006e
non-metallic mineral products industry in the context of a low-
2010 mainly counted on the elimination of backward production
carbon economy.
capacities in high energy-consuming industries. The double indexes
Carbon dioxide emission increase in the Chinese non-metallic
were then introduced by the Chinese government, which stipulated
mineral products industry was mainly due to the industrial activ-
that both energy intensity and carbon intensity would decrease by
ity (IA) effect. As shown in Fig. 3, the increased CO2 emissions
40%e45% in 2020 compared to the level in 2005. The index of carbon
contributed by the IA effect showed an inverted U-shaped trend
intensity encompasses two meanings: first, the modification of en-
during the time intervals. The logic behind this phenomenon is
ergy structure; second, the diversification of fuel mix towards lower
quite clear. With the promotion of urbanization in China, there was
emission fuels. Therefore, the negative impacts of EI on the CO2
an increasing demand of raw materials for the infrastructure con-
emissions increase are expected to be greater during 2011e2020.
struction; therefore, the product output as well as industrial value
added (IVA) grew rapidly during the urbanization process. The
slopes of both IVA and CO2 emissions curves were relatively flat 4.3. Aggregated effects of CO2 emissions change
before the year 2003 (Fig. 4). The annual average growth rate of IVA
was 10.05% during 1985e2002, which pushed up CO2 emissions Some important information might be left out if the influencing
from energy consumption by 2.29% annually. factors are calculated based on time intervals. In order to avoid such
It is worth noting that the rapid growth of IVA due to the problems, we further calculate the aggregated effects of contribu-
accelerated urbanization process after 2002 did not lead to the tors to energy-related CO2 emissions change in non-metallic min-
sharp increase in CO2 emissions. The average annual growth rate of eral products industry in China. The following calculation, which is
IVA reached 18.71% during 2003e2010 while the growth rate of CO2 also based on equations from (4)e(9), uses 1986 as the base year.
emissions was only 9.18% during the same period. It was mainly Results are shown in Fig. 5.
because of the industrial structure adjustment and the decreased As shown in Fig. 5, CO2 emissions change from energy con-
energy intensity. CO2 emissions due to the IA effect increased sumption in the Chinese non-metallic mineral products industry
dramatically in periods 1996e2000 and 2001e2005, which increased from 13.64 Mt in 1987 to 308.48 Mt in 2010, equivalent to
reached 224.34 Mt and 217.44 Mt, respectively. However, the a growth of 2162.40%. Since 2002, carbon dioxide emissions change
decoupling of industrial growth and CO2 emissions was achieved has resumed their growth at an even greater rate [26], because of
during 2006e2010 due to the mandatory reduction of energy in- China’s accelerated process of urbanization. The urbanization pro-
tensity, and the increased CO2 emissions contributed by the IA ef- cess boosted the infrastructure construction, which then promoted
fect further decreased to 155.35 Mt during 2006e2010. the demands for cement and other building materials. Inevitably,
The decreased CO2 emissions contributed by the energy intensity the growth of industrial output will result in the growth of energy
effect (EI) showed an overall upward trend during the time intervals. consumption and the corresponding CO2 emissions. As a result,
Due to the industrial restructuring launched by the Chinese gov- the average annual growth rate of CO2 emissions change during
ernment, the negative impact of EI on CO2 emissions declined during 2003e2010 was 26.28%.
1996e2000. In the process of industrial restructuring, a large The industrial activity (IA) effect was the major factor contrib-
number of private and state-owned companies went bankrupt, uting to the increase in CO2 emissions change (Fig. 5). Compared
leading to the deceleration in growth of output and IVA. with the base year 1986, CO2 emissions increase contributed by IA
The influence of EI on CO2 emissions decrease was remarkable increased from 0.80 Mt in 1987 to 953.36 Mt in 2010. The average
during 2006e2010. It was mainly due to the target in the national annual growth rate of CO2 emissions increase attributed by IA was
694 B. Lin, X. Ouyang / Energy 68 (2014) 688e697

Table 1
Hypothesis of variables (Unit: %).

Scenarios CI S EI IA IS

BAU scenario 0.49 0.32 5.22 13.62 0.29

Scenario 1 0.29 0.12 5.02 13.82 0.09
Scenario 2 0.69 0.52 5.42 13.42 0.49

employment (g) for the forecasting year are given as exogenous, the
prediction of future energy-related CO2 emissions of the Chinese
non-metallic mineral products industry can be made based on
equations (11)e(16).
In this paper, we estimate the future CO2 emissions of the Chi-
nese non-metallic mineral products industry in 2020. The year
2020 is an important time node for China, because the rapid ur-
banization and industrialization process of China is expected to
Fig. 5. Aggregated effects of contributors to energy-related CO2 emissions change in
come to an end in 2020 [47]; more importantly, the modification of
the Chinese non-metallic mineral products industry. both industrial structure and energy consumption structure could
be realized after 2020. In order to evaluate the degree of influence
of each factor on the future CO2 emissions, we set three scenarios to
121.73% during 1986e2010. Meanwhile, the major contributor to the
provide a comprehensive comparison of results. Similar scenario
decline of CO2 emission change was energy intensity (EI) effect. The
analysis has been applied by Ou et al. [48], IEA [49], Zhang et al. [50]
negative contribution increased from 15.25 Mt in 1986 to 486.46 Mt
and so on and so forth.
in 2010. The average annual growth rate of EI effect, which was
The three scenarios are as follows: (i) BAU (business as usual)
lower than that of the IA effect, was 109.22% during 1986e2010.
scenario, of which the set is based on the historical development
Although the impact was small, the substitution (S) effect
trend of each factor. The parameters chosen for BAU in this paper
contributed to the decline of CO2 emissions change over the study
are the average growth rate of each variable (CI, S, EI, IA, IS) during
period. The small effect also illustrated that energy structure of the
Chinese non-metallic mineral products industry was still domi- 1986e2010. A BAU scenario reflects the possible variation trend
indeed; however, the purpose of a BAU scenario is not to provide
nated by coal. The effect of carbon intensity of energy (CI) showed a
varying trend in several years; however, it contributed to the precise estimates of the specific economic conditions but to make
clarifications of the major factors that contribute to the future CO2
decline of CO2 emissions change overall. The decline in CO2 emis-
sions contributed by the CI effect increased from 1.42 Mt in 1987 to emissions. Moreover, it is the benchmark for setting other two
scenarios. (ii) Scenario 1, the higher-emission scenario, of which
49.09 Mt in 2010 compared with the base year 1986.
The effect of industrial scale (IS) was also unstable during the the growth rate of each factor is two percentages higher than that
in the BAU scenario. (iii) Scenario 2, the lower-emission scenario,
study period. However, it was the third most influential factor
among variables, which was worthy of our attention. During 1987e of which the growth rate of each factor is two percentages lower
than that in the BAU scenario. As previously mentioned, the set of
1997, the IS effect contributed to the increase in CO2 emissions
change expect for the year 1990. However, even though the impact other two scenarios are based on the BAU scenario. The advantage
of the scenario analysis is that we can have a relatively compre-
of IS effect on CO2 emissions change was negative, it was quite
small (0.94 Mt). Afterward, the IS effect contributed to the CO2 hensive analysis on the possible results of CO2 emissions, so that
different policies on energy utilization and carbon dioxide emis-
emissions decrease. The negative impacts on CO2 emissions
increased from 104.10 Mt in 1998 to 155.59 Mt in 2005, the peak of sions can be compared. Hypothesis of variables is shown in Table 1.
Based on the hypothesis above, estimates of CO2 emissions from
amount of decreased CO2, and then dropped to 83.47 Mt in 2010.
The negative impacts of IS effect on the CO2 emissions increase may energy consumption in the Chinese non-metallic mineral products
industry in 2020 are shown in Fig. 6. The scenario-differentiated
due to the industrial consolidation and the elimination of backward
production capacity, which were beneficial to energy conservation growth rates of explanatory variables, which consider factors such
as the actual growth conditions, the growth promoted by urbaniza-
and energy use reduction.
In summary, the major contributors to CO2 emissions change in tion and the possible adaptations to the low-carbon development in
the Chinese non-metallic mineral products industry, are reasonable.
the Chinese non-metallic mineral products industry were factors of
industrial activity (IA) and energy intensity (IE). The effect of in- As shown in Fig. 6, CO2 emissions from energy consumption in
the Chinese building materials industry will keep increasing. It can
dustrial scale (IS), which varied during our study period, indicated
be explained by two reasons: first, industrial energy demand will
significant negative impacts on the CO2 emissions increase during
keep growing in the urbanization and industrialization stage [9,51];
1998e2010. Although the impacts were quite small, the effects of
second, industrial energy consumption structure will remain stable
carbon intensity of energy (CI) and substitution (S) also contributed
during the rapid economic development process [7].
to the decrease of CO2 emissions.
Under the BAU scenario, CO2 emissions from energy consump-
tion of the Chinese non-metallic mineral products industry in 2020
4.4. Emission reduction potential
are predicted to be 953.22 Mt, accounting for 10.12% of China’s total
CO2 emissions.2 In addition, CO2 emissions from energy
4.4.1. Estimates of energy-related CO2 emissions
According to equation (3), CO2(0) ¼ CI(0)*S(0)*EI(0)*IA(0)*IS(0)
for the base year 2011 can be calculated If the change rates of the 2
The share of CO2 emissions from energy consumption of the Chinese non-
CO2 emissions per unit of fossil-fuel (a), the share of fossil fuels in metallic mineral products industry in China’s total CO2 emissions is based on the
total energy (b), energy consumption per unit of industrial value estimate from Lin and Liu [51], which shows that China’s CO2 emissions will reach
added (d), industrial value added per capita (f) and the 9.418 billion tons in 2020 under the baseline scenario.
 B. Lin, X. Ouyang / Energy 68 (2014) 688e697 695

Table 2
Reduction potential of CO2 emissions in the Chinese non-metallic mineral products
industry and its impact on China’s total CO2 emissions.

Year Scenario of moderate emission Scenario of aggressive emission

reduction reduction

CO2 emissions reduction Share CO2 emissions reduction Share

2015 35.22 N.A. 68.82 N.A.

2020 99.02 1.05 188.88 2.01

Note: N.A. stands for not available.

5. Conclusions and policy suggestions

This article assesses the CO2 emissions change from energy

consumption in the Chinese non-metallic mineral products in-
Fig. 6. Actual values and estimates of energy-related CO2 emissions in the Chinese
non-metallic mineral products industry under different scenarios.
dustry based on the application of Logarithmic Mean Divisia Index
(LMDI) method. The results show that industrial activity (IA) effect
was the leading force for the increase of CO2 emissions while en-
ergy intensity (EI) effect was the major contributor for the CO2
consumption will reach 1052.24 Mt and 863.35 Mt in 2020 under
emissions decrease, which is in agreement with [53e55]. It was
scenario 1 and scenario 2, respectively.
observed that fuel diversification (S) effect and carbon intensity of
energy (CI) are amongst the factors that contributed to the decline
4.4.2. Estimates of emission reduction potential of CO2 emissions. Industrial scale (IS) effect presented a varying
In order to estimate the future reduction potential of CO2 weight interchanging intervals of growth.
emissions in the Chinese non-metallic mineral products industry, To the best of our knowledge, this study is evaluated for the first
we further develop two scenarios: (i) scenario of moderate emis- time in the field of CO2 emissions change from energy consumption
sion reduction, (ii) scenario of aggressive emission reduction. The in the Chinese non-metallic mineral products industry. Results in
above set is based on three reasons: first, it will be difficult for the this article offer complementary perspectives on determinants of
Chinese non-metallic mineral products industry to reduce CO2 energy-related CO2 emissions change and help to formulate miti-
emissions considering the increasing demand of raw materials for gation strategies for CO2 emissions [56] in industrial sectors in
infrastructure construction; second, the extensive growth of in- China. Results demonstrate that reduction potentials of CO2 emis-
dustry must be supported by lower cost of energy; and coal is the sions are 99.02 Mt and 188.88 Mt in 2020 under the scenarios of
most competitive energy due to relative abundant reserves. How- moderate and aggressive emission reduction, respectively, which
ever, it is the energy with the highest carbon content; third, the correspondingly account for 0.30% and 0.58% of the world’s energy-
reduction targets of energy intensity and carbon intensity in 2020 related CO2 emissions in 2011.
will promote clean energy development and energy efficiency Measures for CO2 emissions reduction are suggested as follows:
improvement in the industrial sector. Therefore, potentials of CO2 Energy conservation should be the major strategy for reducing
emissions reduction are influenced by the industrial development energy intensity [57]. Compared to energy structural adjustment,
as well as government policy. energy conservation is a cheaper and more feasible way to reduce
Scenarios of emission reduction are set as follows: under the energy consumption in industrial sectors. For the Chinese non-
scenario of moderate emission reduction, CO2 emissions of the metallic mineral products industry, energy conservation could be
Chinese non-metallic mineral products industry will transform achieved by several measures: first, adjusting the product structure
from Scenario 1 to BAU scenario. Accordingly, under the scenario of by reducing the production of high energy-consuming products;
aggressive emission reduction, CO2 emissions of the Chinese non- second, closing down backward production capacities; third,
metallic mineral products industry will transform from Scenario substituting energy with capital investment (i.e., upgrading new
1 to Scenario 2. Similar set is applied by Lin and Ouyang [52]. Ac- equipment) in the production process. EMC (Energy management
cording to estimates in Section 4.4.1, reduction potential of CO2 contracting), which has been widely adopted in industrial energy
emissions is calculated and reported in Table 2. conservation and is quite popular in developed countries, is also an
As shown in Table 2, reduction potential of CO2 emissions in important way to improve energy utilization efficiency. The gov-
the Chinese non-metallic mineral products industry is substan- ernment should promote the EMC in China’s industrial sectors by
tial. Under the scenario of moderate emission reduction, overcoming systematical obstacles, such as the recognition of en-
99.02 Mt of CO2 emissions can be reduced in 2020, which is ergy service corporations and the financing of energy-saving
equivalent to the energy-related CO2 emissions of Greece in 2009 management. Furthermore, the government could encourage in-
(99.83 Mt) and accounts for 0.30% of the world’s energy-related vestment on energy saving technology by providing tax exemp-
CO2 emissions in 2011. Moreover, the corresponding share of tions and preferential loans [58].
CO2 emissions reduction in China’s total CO2 emissions would be Energy pricing reform is a long-term strategy for CO2 emissions
1.05% in 2020. reduction. Fuel diversification and cleaner energy structure are
Under the scenario of aggressive emission reduction, reduction effective ways to constrain the increase of CO2 emissions. In order
potential of CO2 emissions in the Chinese non-metallic mineral to modify the coal-dominated energy structure, the consumption of
products industry is predicted to be 188.88 Mt in 2020, which is coal and other fossil fuels must be replaced by cleaner energy.
close to the energy-related CO2 emissions of Egypt in 2010 However, industrial sectors will have little incentives to adjust their
(190.59 Mt) and accounts for 0.58% of the world’s CO2 emissions energy structures if the government kept subsidizing the con-
from energy consumption in 2011. In addition, the corresponding sumption of fossil fuels. Only if the true costs of fossil fuels are
share of CO2 emissions reduction in China’s total CO2 emissions reflected in energy prices, the renewable energy will be competi-
would be 2.01% in 2020. tive. Moreover, the Chinese government should further promote
696 B. Lin, X. Ouyang / Energy 68 (2014) 688e697

the reform of electricity pricing because most renewable energies [22] Ang BW, Liu N. Handling zero values in the logarithmic mean Divisia index
decomposition approach. Energy Policy 2007;35(1):238e46.
are used for power generation. Considering the environmental
[23] Ang BW, Huang HC, Mu AR. Properties and linkages of some index decom-
advantages and positive externalities [59], increasing the subsidy position analysis methods. Energy Policy 2009;37(11):4624e32.
for renewable energy is also a common scheme used by many [24] Wang C, Chen J, Zou J. Decomposition of energy-related CO2 emission in
countries in solving the high-cost problem of renewable energy. To China: 1957e2000. Energy 2005;30(1):73e83.
[25] Liu LC, Fan Y, Wu G, Wei YM. Using LMDI method to analyze the change of
summarize, energy pricing reform is the essential method to China’s industrial CO2 emissions from final fuel use: an empirical analysis.
reduce CO2 emissions, which can also effectively reduce the Energy Policy 2007;35(11):5892e900.
rebound effect of energy consumption due to energy efficiency [26] Ma C, Stern DI. Biomass and China’s carbon emissions: a missing piece of
carbon decomposition. Energy Policy 2008;36(7):2517e26.
improvement [60]. Experiences from developed countries [27] Malla S. CO2 emissions from electricity generation in seven AsiaePacific and
demonstrate that market-orientated methods for energy conser- North American countries: a decomposition analysis. Energy Policy 2009;37(1):
vation and emission reduction are more effective than government 1e9.
[28] Zhao M, Tan L, Zhang W, Ji M, Liu Y, Yu L. Decomposing the influencing factors
interventions [61,62]. of industrial carbon emissions in Shanghai using the LMDI method. Energy
2010;35(6):2505e10.
[29] Zha DL, Zhou DQ, Zhou P. Driving forces of residential CO2 emissions in urban
Acknowledgments and rural China: an index decomposition analysis. Energy Policy 2010;38(7):
3377e83.
[30] Oh I, Wehrmeyer W, Mulugetta Y. Decomposition analysis and mitigation
The authors appreciate the editor and two anonymous reviewers strategies of CO2 emissions from energy consumption in South Korea. Energy
for their invaluable and constructive comments. The paper is sup- Policy 2010;38(1):364e77.
[31] Akbostancı E, Tunç GI, _ Türüt-Aşık S. CO2 emissions of Turkish manufacturing
ported by Newhuadu Business School Research Fund, Ministry of
industry: a decomposition analysis. Appl Energy 2011;88(6):2273e8.
Education Foundation (Funding no. 10JZD0018), Basic Research
[32] Zhang Y, Zhang J, Yang Z, Li S. Regional differences in the factors that influence
Universities Special Foundation (Funding no. 2010221051), Ministry China’s energy-related carbon emissions, and potential mitigation strategies.
of Education Foundation (Funding no. 10JBG013), and National Energy Policy 2011;39(12):7712e8.
[33] Liu Z, Geng Y, Lindner S, Guan D. Uncovering China’s greenhouse gas emission
Social Science Foundation (Funding no. 09&ZD050).
from regional and sectoral perspectives. Energy 2012;45(1):1059e68.
[34] Hammond GP, Norman JB. Decomposition analysis of energy-related carbon
emissions from UK manufacturing. Energy 2012;41(1):220e7.
References [35] Fujii H, Managi S, Kaneko S. Decomposition analysis of air pollution abate-
ment in China: empirical study for ten industrial sectors from 1998 to 2009.
[1] Chen W, Wu Z, He J, Gao P, Xu S. Carbon emission control strategies for China: J Clean Prod 2013;59(15):22e31.
a comparative study with partial and general equilibrium versions of the [36] Shao C, Guan Y, Wan Z, Guo C, Chu C, Ju M. Performance and decomposition
China MARKAL model. Energy 2007;32(1):59e72. analyses of carbon emissions from industrial energy consumption in Tianjin,
[2] Lin B, Sun C. Evaluating carbon dioxide emissions in international trade of China. J Clean Prod 2014;64:590e601.
China. Energy Policy 2010;38(1):613e21. [37] Bacon RW, Bhattacharya S. Growth and CO2 emissions: how do different
[3] Liu LC, Wu G, Wang JN, Wei YM. China’s carbon emissions from urban and countries fare?In Environment department papers, climate change series. The
rural households during 1992e2007. J Clean Prod 2011;19(15):754e1762. World Bank Environment Department; 2007. paper number 113. Available
[4] Price L, Sinton J, Worrell E, Phylipsen D, Xiulian H, Ji L. Energy use and carbon from: http://siteresources.worldbank.org/INTCC/214574-1192124923600/
dioxide emissions from steel production in China. Energy 2002;27(5):429e46. 21511758/CO2DecompositionfinalOct2007.pdf [accessed 22.07.13].
[5] EIA. Total carbon dioxide emissions from the consumption of energy (million [38] Kaya Y. Impact of carbon dioxide emission on GNP growth: interpretation of
metric tons). International Energy Statistics; 2013. Available from: http:// proposed scenarios. In: Presentation to the energy and industry subgroup,
www.eia.gov/cfapps/ipdbproject/iedindex3.cfm?tid¼90&pid¼44&aid¼8&cid¼ response strategies working group. Paris: IPCC; 1989.
regions&syid¼2000&eyid¼2011&unit¼MMTCD [accessed 22.10.13]. [39] China statistical yearbook. Beijing: China Statistics Press; 1986e2011 [in
[6] EIA. Total carbon dioxide emissions from the consumption of Energy. Avail- Chinese].
able from: http://www.eia.gov/cfapps/ipdbproject/iedindex3.cfm?tid¼90& [40] China energy statistical yearbook. Beijing: China Statistics Press; 1986e2011
pid¼44&aid¼8&cid¼CH,&syid¼1980&eyid¼2011&unit¼MMTCD; 2013 [accessed [in Chinese].
01.07.13]. [41] IPCC. IPCC guidelines for national greenhouse gas inventories. In:
[7] Lin B, Moubarak M. Decomposition analysis: change of carbon dioxide Eggleston HS, Buendia L, Miwa K, Ngara T, Tanabe K, editors. Prepared by the
emissions in the Chinese textile industry. Renew Sustain Energy Rev 2013;26: national greenhouse gas inventories programme. Japan: IGES; 2006.
389e96. [42] CEIC China database. Available from: http://ceicdata.securities.com/cdmWeb/;
[8] Lin BQ, Liu JH. Estimating coal production peak and trends of coal imports in 2013 [accessed 18.08.13].
China. Energy Policy 2010;38(1):512e9. [43] CEInet industry database. Available from: http://ceidata.cei.gov.cn/aspx/
[9] Lin BQ, Ouyang XL. Energy demand in China: comparison of characteristics Default.aspx; 2013 [accessed 16.09.13].
between the US and China in rapid urbanization stage. Energy Convers [44] Lin BQ. The econometric research on Energy demands of China. J Stat Res
Manage 2014;79:128e39. 2001;10:34e7 [in Chinese].
[10] Ang BW, Zhang FQ. A survey of index decomposition analysis in energy and [45] NDRC, National Development and Reform Commission. Speeding up major
environmental studies. Energy 2000;25(12):1149e76. infrastructure construction (in Chinese). Available from:, http://www.ndrc.
[11] Ang BW. Decomposition analysis for policymaking in energy: which is the gov.cn/xwfb/t20130804_552752.htm; 2013 [accessed 16.09.13].
preferred method? Energy Policy 2004;32(9):1131e9. [46] Ang BW. Is the energy intensity a less useful indicator than the carbon factor
[12] Ang BW, Pandiyan G. Decomposition of energy-induced CO2 emissions in in the study of climate change? Energy Policy 1999;27(15):943e6.
manufacturing. Energy Econ 1997;19(3):363e74. [47] The World Bank. China 2030: building a modern, harmonious, and creative
[13] Choi KH, Ang BW. A time-series analysis of energy-related carbon emissions high-income society. Washington D.C., U.S.A; 2012.
in Korea. Energy Policy 2001;29(13):1155e61. [48] Ou X, Zhang X, Chang S. Scenario analysis on alternative fuel/vehicle for
[14] Sun J. Changes in energy consumption and energy intensity: a complete China’s future road transport: life-cycle energy demand and GHG emissions.
decomposition model. Energy Econ 1998;20(1):85e100. Energy Policy 2010;38(8):3943e56.
[15] Ang BW, Choi KH. Decomposition of aggregate energy and gas emission in- [49] IEA. World energy outlook 2012. Paris: OECD; 2012.
tensities for industry: a refined Divisia index method. Energy J 1997;3:59e73. [50] Zhang Q, Ishihara KN, Mclellan BC, Tezuka T. Scenario analysis on future
[16] Ang BW, Zhang FQ, Choi KH. Factorizing changes in energy and environmental electricity supply and demand in Japan. Energy 2012;38(1):376e85.
indicators through decomposition. Energy 1998;23(6):489e95. [51] Lin BQ, Liu XY. China’s carbon dioxide emissions under the urbanization
[17] Ang BW, Zhang FQ. Inter-regional comparisons of energy-related CO2 emis- process: influence factors and abatement policies. J Econ Res 2010;8:66e78
sions using the decomposition technique. Energy 1999;24(4):297e305. [in Chinese].
[18] Zhang FQ, Ang BW. Methodological issues in cross-country/region decom- [52] Lin BQ, Ouyang XL. Electricity demand and conservation potential in the Chi-
position of energy and environment indicators. Energy Econ 2001;23(2): nese nonmetallic mineral products industry. Energy Policy [accepted paper].
179e90. [53] Alves MR, Moutinho V. Decomposition analysis and innovative accounting
[19] Ang BW. The LMDI approach to decomposition analysis: a practical guide. approach for energy-related CO2 (carbon dioxide) emissions intensity over
Energy Policy 2005;33(7):867e71. 1996e2009 in Portugal. Energy 2013;57(1):775e87.
[20] Ang BW, Liu FL. A new energy decomposition method: perfect in decompo- [54] Inglesi-Lotz R, Pouris A. Energy efficiency in South Africa: a decomposition
sition and consistent in aggregation. Energy 2001;26(6):537e48. exercise. Energy 2012;42(1):113e20.
[21] Ang BW, Liu FL, Chew EP. Perfect decomposition techniques in energy and [55] Kumbarog lu G. A sectoral decomposition analysis of Turkish CO2 emissions
environmental analysis. Energy Policy 2003;31(14):1561e6. over 1990e2007. Energy 2011;36(5):2419e33.
 B. Lin, X. Ouyang / Energy 68 (2014) 688e697 697

[56] Andreoni V, Galmarini S. European CO2 emission trends: a decomposition [60] Lin B, Yang F, Liu X. A study of the rebound effect on China’s current energy
analysis for water and aviation transport sectors. Energy 2012;45(1):595e602. conservation and emissions reduction: measures and policy choices. Energy
[57] He J, Deng J, Su M. CO2 emission from China’s energy sector and strategy for its 2013;58:330e9.
control. Energy 2010;35(11):4494e8. [61] Li A, Zhang A, Cai H, Li X, Peng S. How large are the impacts of carbon-
[58] Ates SA, Durakbasa NM. Evaluation of corporate energy management prac- motivated boarder tax adjustments on China and how to mitigate them?
tices of energy intensive industries in Turkey. Energy 2012;45(1):81e91. Energy Policy 2013;63:927e34.
[59] Lund H. Renewable energy strategies for sustainable development. Energy [62] Li A, Lin B. Comparing climate policies to reduce carbon emissions in China.
2007;32(6):912e9. Energy Policy 2013;60:667e74.