Peak energy consumption and CO2 emissions in China

Energy Policy 68 (2014) 508–523

Contents lists available at ScienceDirect

Energy Policy journal homepage: www.elsevier.com/locate/enpol

Peak energy consumption and CO2 emissions in China Jiahai Yuan a,n, Yan Xu a, Zheng Hu b,nn, Changhong Zhao a, Minpeng Xiong a, Jingsheng Guo a a b

School of Economics and Management, North China Electric Power University, Beijing, China Center for Energy and Environmental Policy, University of Delaware, DE, USA

H I G H L I G H T S








A framework for modeling China's energy and CO2 emissions is proposed. Scenarios are constructed based on various assumptions on the driving forces. Energy consumption will peak in 2035–2040 at 5200–5400 Mtce. CO2 emissions will peak in 2030–2035 at about 9300 Mt and be cut by 300 Mt in a cleaner energy path. Energy consumption and CO2 emissions per capita will peak soon after China steps into the high income group.

art ic l e i nf o

a b s t r a c t

Article history: Received 14 November 2013 Received in revised form 27 December 2013 Accepted 16 January 2014 Available online 4 February 2014

China is in the processes of rapid industrialization and urbanization. Based on the Kaya identity, this paper proposes an analytical framework for various energy scenarios that explicitly simulates China's economic development, with a prospective consideration on the impacts of urbanization and income distribution. With the framework, China's 2050 energy consumption and associated CO2 reduction scenarios are constructed. Main findings are: (1) energy consumption will peak at 5200–5400 million tons coal equivalent (Mtce) in 2035–2040; (2) CO2 emissions will peak at 9200–9400 million tons (Mt) in 2030–2035, whilst it can be potentially reduced by 200–300 Mt; (3) China's per capita energy consumption and per capita CO2 emission are projected to peak at 4 tce and 6.8 t respectively in 2020–2030, soon after China steps into the high income group. & 2014 Elsevier Ltd. All rights reserved.

Keywords: Energy consumption CO2 emissions long-term scenario China

1. Introduction The pattern of energy consumption plays an essential role in social-economic development. The growth of per capita energy consumption and per capita CO2 emissions shows an approximately inverted “U” curve, namely Environmental Kuznets Curve (EKC) (Selden and Song, 1994; Grossman and Krueger, 1995). There are debates on whether EKC could be found in developed countries' CO2 emissions/energy consumption pattern (Iwata et al., 2011; Jakob et al., 2012). The global achievement of ambitious climate targets particularly requires radical reduction of CO2 emissions in industrialized countries as well as control of emissions in developing countries (IPCC, 2007). However, there have been opposite trends, for instance, Raupach et al. (2007) found that in 2000– 2004, economic growth in developing and less developed countries were the main driver for increasing global CO2 emissions. China became globally the largest CO2 emitter since 2009. The Chinese government proposed to address the emission issue by releasing several energy targets: 20% reduction of energy intensity n

Corresponding author. Tel.: þ 86 10 61773 091. Corresponding author. E-mail addresses: [email protected] (J. Yuan), [email protected] (Z. Hu).

nn

0301-4215/$ - see front matter & 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.enpol.2014.01.019

of the economy by 20% during 2005–2010, 40–45% reduction of GDP CO2 intensity during 2005–2020, and a 15% clean energy share target by 2020 (Yuan et al., 2011; State Council of China, 2013). In the 18th National Congress of CPC, China's new administrative government released stronger targets of doubling total GDP and per capita income by 2020 on the 2010 levels (Hu, 2012). Meanwhile, the concept of “ecological civilization and beautiful China” was firstly proposed, which indicates Chinese Government's high attention to the environmental and ecological impact of energy consumption. China is currently in the process of industrialization and urbanization. The following questions are proposed: “what is the expected growth of China's future energy consumption?”, “would CO2 emissions reach a peak (or a plateau) or would it be monotonously increase in the long run?”, and “when and what is the peak of China's CO2 emission?” This paper seeks to address these questions, from the perspective of China's recently-declared social-economic policies and related energy and environmental impacts. The paper is organized as follows: Section 2 reviews literatures of current discussions. Section 3 presents the methodology. Section 4 provides the historical trajectory of driving factors and sets assumptions for the scenarios. Section 5 presents the

J. Yuan et al. / Energy Policy 68 (2014) 508–523

scenarios from the perspective of international comparative study. Section 6 concludes the paper.

2. Literature review 2.1. Environmental Kuznets curve The EKC hypothesis implies environmental issues would be concerned only when economic growth reaches a certain level (Grossman and Krueger, 1991; IBRD, 1992). According to Stern (2004), The EKC seeks to expound and prove a two-track argument: (1) whether there is an inflection point in the curve of per capita CO2 emissions vs. GDP per capita; and (2) whether there is a reflection of convergence on per capita CO2 emissions. Regarding inflection point, Richmond and Kaufmann (2006) found limited evidence of EKC in OECD countries. Jaunky (2011) found no inflection point in 36 high-income countries for the period 1980–2005 but indicated that over time CO2 emissions would be stable. For developing countries, no inflection point has been found, which is consistent with the EKC hypothesis (Jakob et al., 2012; Jaunky, 2011). Regarding convergence, Some scholars argued that CO2 emissions of the OECD countries showed a significant sign of convergence, according to Strazicich and List (2003), Aldy (2006), Lee and Chang (2008), Romero-Ávila (2008), Westerlund and Basher (2008), and Meng et al. (2013); while on the other hand, Barassi et al. (2008) suggested an opposite argument. Van (2005), Stegman (2005), Panopoulou and Pantelidis (2009), and Jobert et al. (2010) found strong evidence of global convergence; however, some others proved the divergence of per capita CO2 emissions; e.g., Aldy (2006). Lin and Li (2013) found an absolute convergence of per capita CO2 emissions over the period of 1971–2008 within subsamples grouped by income level. Furthermore, there was an argument of lacking evidence of absolute convergence in the full sample containing 110 countries. Methods of data and variable selection are key determinations of various EKC results. For instance, due to changes in fuel composition's direct impacts on emissions, per capita energy consumption is more appropriate than CO2 emissions per capita as the dependent variable. But for a sample of 113 countries covering the period 1971–2004, no evidence of EKC was detected in the world as a whole or in single country level (Luzzati and Orsini, 2009). However, Jakob et al. (2012) found the existence of EKC in the panel data of 21 industrialized countries over the period of 1971–2005. Sun (1999) argued that CO2 EKC merely reflects the theory of peak-value for energy intensity. What is behind is the structure shifts from the higher energy intensity of heavy industry to the lower intensity of light industry; and the product structure changes from general value-added to higher value-added, from material production to knowledge production. Therefore, we believe that CO2 EKC is a reflection of the historical pattern of energy intensity; it is not the guidance to determine when a country's environment starts to improve. What's interesting is that inflection points of both energy intensity curve and CO2 EKC occurred in China in 1977, where per capita GDP was 250US$. Jalil and Mahmud (2009) also found the evidence of EKC for China during 1975–2005.

consumption and GHG emissions in 2000–2050. Accordingly, energy consumption will continue to increase, whilst under the reinforcement of low-carbon scenario, CO2 emissions is estimated to reach the peak in 2030. Lin and Jiang (2009) applied the original CO2 EKC simulation model to predict China's emission and found that inflection point is likely to happen in 2020. In addition, International Energy Agency (IEA) set an aggressive “450 Scenario” by 2030 with China-specific policy assumptions and outlook (IEA, 2009a). Lawrence Berkeley National Laboratory (LBNL) projects China's primary energy demand would rise continuously and approach a plateau around 2040. CO2 emissions will approach a plateau in 2025 or 2030 depending on the underlying assumptions. Energy Research Institute (ERI) of National Development and Reform Commission (NDRC) published a 2050 China Energy and CO2 Emissions Report in2009, which described the potential energy and emissions scenarios in 2050 (CEACER, 2009), where energy demand in 2050 is expected to reach 6690 Mtce in the baseline scenario and 5560 Mtce in the low-carbon scenario. Moreover, Chinese Academy of Engineering proposed a “scientific, green and low-carbon” energy strategy and projecting CO2 emission to reach 9 billion tones in 2030 (CAE, 2011). To sum up, firstly, EKC can be used as an important tool to observe the income-pollutant relationship for developing countries. Secondly, due to the complexity of CO2 emissions, some additional factors would be essential to model CO2 emissions. Thirdly, complexity of modeling structuring and variables would result in difficulty of understanding scenarios. Furthermore, subjective assumption of parameter/scenario settings could lead to incommensurable results and misleading policy.

3. Methodology of the study The Kaya identity states that total emission level can be expressed as the product of four inputs: population, GDP per capita, energy use per unit of GDP, and carbon emissions per unit of consumed energy (Kaya and Yokobori, 1997). In order to adopt the Kaya identity to capture China's characteristics, firstly, residential energy consumption accounts for only 10% of primary energy but the share would be substantially increasing. It is important to differentiate energy consumption for production and for daily life. Secondly, due to the severe gap between rural and urban household energy consumption, it is essential to model the impact of urbanization process. Thirdly, since China's economic structure is transmitting to service-based economy and China's energy efficiency is still low, we want to model the impact of economic output as well as energy efficiency improvements. Accordingly, we adapt the equation as follows: EI GDP ¼

E EP þ E R ¼ GDP GDP

EP EPP þ EPS þ EPS ¼ ¼ ∑ ðSi I i Þ GDP GDP i ¼ 13

CI GDP ¼

Various scenarios show that China's future energy consumption and GHG emissions have been largely improved (Vuuren van, et al., 2003; Wang and Waston, 2009, 2010; McKinsey and Company, 2009; Steckel et al., 2011; Rout et al., 2011; Zhou et al., 2013). Jiang et al. (2009) analyzed China's long-term scenarios of energy

ð1Þ ð2Þ

eU IU eR IR  P  RU þ   P  ð1  RU Þ  I U gdp I R gdp

ð3Þ

CO2 Emissions E CO2 Emissions ¼  ¼ EI GDP  EF E GDP GDP E

ð4Þ

ER ¼ EUR þ ERR ¼

EF E ¼ ∑ðEs ef Þ 2.2. Peak energy consumption and CO2 emissions in China

509

ð5Þ

i

CO2 emissions ¼ GDP  CI GDP

ð6Þ

According to the model, primary energy consumption (E) is decomposed into production use (EP) and household use (ER). The productive factor can be expressed as the sum of products of all output shares (primary, secondary and tertiary, Si) and their corresponding energy

510

J. Yuan et al. / Energy Policy 68 (2014) 508–523

intensities (Ii). The residential factor can be distinguished by two scenarios, rural and urban consumption. Hence, the process of urbanization (RU) in China and its impact would be represented. Energy consumption is an essential part of residential consumption, which in turn, is subject to the status of disposable income (IU) for rural net per capita income and IR for urban disposable per capita income. In China, income growth was slower than GDP growth and the government's commitment of improving future income distribution is uncertain. In order to model the impact of income distribution policy on energy consumption, we add two factors of disposable income per capita as an intermediate variable between per capita GDP, and also residential energy consumption per capita. Primary energy CO2 emissions factor (EFE) is the sum of product of fuel emission factors (ef) and their shares in primary energy consumption (ES). Based on Eq. (5), the impact of primary energy structure or different energy development paths on CO2 emissions can be captured.

4. Driving factors of energy-related CO2 emissions in China 4.1. Population 4.1.1. Total fertility rate Total fertility rate (TFR) in China was 2.24 in 1980 and 1.22 in 2000 (NSBC, 2001). It was 1.21 in 2010, which was less than half of the world average, and was also significantly lower than the average of developed countries (Fig. 1) (NSBC, 2011). According to the estimates of Population Reference Bureau and US Census Bureau International Programs (State Council of China, 2011), the

actual TFR in China probably ranges between 1.4 and 1.6. According to the experiences of South Korea, Japan, Singapore and Russia (UN, 2012), it is worth noting that the downtrend of TFR can hardly be reversed. 4.1.2. Population of China during 1980–2011 China's population was 987 million in 1980 and it was 1347 million in 2011 (NSBC, 2012), at a net growth rate of 36.5% with 1.14% average annual growth rate. Although the total population is increasing, a declining trend is shown in the growth rate (Fig. 2). 4.1.3. Projection of total population in China, 2010–2050 In 2011, State Council issued the development plan for population as part of the 12th FYP (State Council of China, 2011). According to this plan, annual growth rate will be managed at 7.2‰ and total population in 2015 will be controlled within 1.39 billion. With the advent of aging society and low fertility rate, the momentum of population growth is very weak in China. The total population is expected to decline based on the long term influence of population planning. A recently revised projection by the United Nations estimated China's population dynamic, which indicated China’s population may reach the peak in 2026 at 1.396 billion. It will decrease to 1.393 billion in 2030 and to 1.296 billion in 2050 (UN, 2012). This projection is used in the population scenario in our study (Fig. 3). 4.2. GDP 4.2.1. GDP growth in China, 1980–2011 After the implementation of reform-and-open-up policy in 1978, China's economic take-off of 10% annual growth has been successful (Fig. 4), comparing with the global average of 3.9% (NSBC, 2012). In 2010, China became the second largest economy with a total GDP of 5878.6 billion US$ (2012-year exchange rate, and hereafter).

2.50

2.00

1.50

1.00

0.50

0.00 1990

1995

2000

2005

2010

1.60

20

1.40

15

1.20 10

1.00 0.80

5

0.60

0

0.40 -5

0.20 0.00 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

-10

Population

Growth Rate

Fig. 2. Total population and its growth in China, 1980–2011.

(‰)

1985

Fig. 1. Total fertility rate in China, 1980–2010.

(Billion)

1980

4.2.2. Per capita GDP growth in China, 1980–2011 According to gross national income (GNI) per capita, an economy is usually classified into three categories: low income, middle income (subdivided into lower middle and upper middle), or high income (WBG, 2012a). According to the 2012 criterion, 1035US$ is the threshold between low income and lower middle income group, 4085US$ is the threshold between low middle and upper middle income group, while 12615US$ is the threshold between middle income and high income group.

J. Yuan et al. / Energy Policy 68 (2014) 508–523

511

6

1.42 1.40

4

1.38 2 1.34

0

1.32

-2

(‰)

(Billion)

1.36

1.30 -4 1.28 -6

1.26

-8

1.24 2010

2015

2020

2025 Population

2030

2035

2040

2045

2050

Growth Rate

40000

16

35000

14

30000

12

25000

10

20000

8

15000

6

10000

4

5000

2

0

(%)

(Billion 2005 RMB)

Fig. 3. Projection of total population and growth rate in China, 2011–2050.

0 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 GDP

Growth Rate

Fig. 4. GDP and its growth in China, 1980–2011.

In 1978, per capita GDP of China was only 155US$. GDP per capita reached 1042US$ in 2001, which indicated China entered into low middle income group. In 2010 per capita GDP reached 4000US$, when China stepped into the upper middle income group (State Council of China, 2011).

4.2.3. Projection of GDP growth in China, 2010–2050 In the report of 18th National Congress of CPC, completion of a well-off society in 2020 on a base of balanced, coordinated and sustainable growth is proposed as a medium target of China's economic vision. GDP and per capita GDP are expected to double by 2020 as of 2010 levels. A long term vision is to reach the income level of moderately developed countries in the middle of 21st century and this vision is interpreted as the current income level of OECD average. Thus, we constructed a Baseline GDP growth scenario to evaluate this assumption. To fully understand the impact of GDP growth on energy consumption, High and Low scenarios are also provided (Table 1). Per capita GDP will reach 5235US$ in 2015 and 7250US$ in 2020 in the baseline scenario. In 2030 it would enter high income group and reach 12600US$, and it would reach 28300US$ in 2050. Note that although the long-term vision is largely realized, double income by 2020 on 2010 levels is not fully achieved in this

scenario. However, in the High scenario, both medium and long term targets can be attained (Fig. 5). 4.3. Industrial structure 4.3.1. Evolution of China's industrial structure, 1980–2011 China's economic growth is accompanied by gradual evolution of industrial structure. Since China is in a transition from the middle to the later stage of industrialization, secondary industry was and still is the leading industry in China's economic structure (Fig. 6) (NSBC, 2012). 4.3.2. Projection of China's industrial structure, 2010–2050 Transforming development pattern requires radical economic restructuring. Urbanization could largely accelerate investment and domestic consumptions, which in turn will create huge demand for secondary industry. Therefore, the share of secondary will not change much in the coming decade. Meanwhile, tertiary industry will experience a boom along with urbanization. In 2010, service industry accounted for 43% of China's total GDP, which was 30–40 percent lower than that in developed countries, and 20 percent lower than the world average (WBG, 2012b). To capture the restructuring impact on energy consumption, two scenarios are provided in the study. The baseline assumes that global average is 3% for primary industry,

512

J. Yuan et al. / Energy Policy 68 (2014) 508–523

Table 1 China's GDP growth scenarios, 2010–2050. Scenario

2010–2015

2015–2020

2020–2025

2025–2030

2030–2035

2035–2040

2040–2045

2045–2050

High (%) Baseline (%) Low (%)

8.0 7.5 6.5

7.5 7.0 6.0

7.0 6.5 5.5

6.5 6.0 5.0

6.0 5.5 4.5

5.5 5.0 4.0

5.0 4.5 3.5

4.5 4.0 3.0

40000 GDP per capita-High Scenario 35000 GDP per capita-Baseline Scenario 30000 (2012 US$)

GDP per capita-Low Scenario 25000 20000 15000 10000 5000 0

2010

2015

2020

2025

2030

2035

2040

2045

2050

Fig. 5. Per capita GDP of China in alternative scenarios, 2010–2050.

100 90 80 70

(%)

60 50 40 30 20

0

1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

10

Primary Industry

Secondary Industry

Tertiary Industry

Fig. 6. China's industrial structure, 1980–2011.

which is lower than 1.5% in developed countries, such as US and Japan. Experience of industrialized countries shows that when GDP per capita reaches 10,000US$ (WBG, 2012b) primary industry will stay at about 5%, secondary at about 40% and tertiary at about 55%. We project that secondary industry will experience its saturation in 2020–2030 and then China would step into the later stage of industrialization (Table 2).

4.4. Urbanization 4.4.1. China's urbanization during 1980–2011 The experience of industrialization demonstrates that China with keep on its social-economic development and there would be a rapid process of urbanization until China reaches a stabilized level. In 1980–2011, urbanization rate increased from 19.39% to 51.27% (Fig. 7) (NSBC, 2012) and the process was accelerated since the middle of 1990s. In 2011 urban population outnumbered rural population, which indicates that China has stepped into the medium-and-later stage of urbanization.

Table 2 Projection of China's industrial structure, 2010–2050. Year

Primary

Secondary

Tertiary

Baseline Acceleration Baseline Acceleration Baseline Acceleration 2010 2015 2020 2025 2030 2035 2040 2045 2050

10.10% 8.10 6.10 5.60 5.10 4.60 4.10 3.60 3.00

7.60 5.10 4.10 3.10 2.70 2.30 1.90 1.50

46.67% 46.17 45.67 45.17 44.67 41.67 38.67 35.67 33.00

45.17 43.67 42.67 41.67 37.50 33.33 29.17 25.00

43.24% 45.74 48.24 49.24 50.24 53.74 57.24 60.74 64.00

46.74 50.24 52.74 55.24 59.80 64.37 68.93 73.50

4.4.2. Projection of China's urbanization, 2010–2050 Experience shows that 50% is an inflection point of urbanization for many countries. Improving the quality and strengthening urban management should be the key task of urbanization after this inflection point. According to NDRC (2013), the speed of urbanization

J. Yuan et al. / Energy Policy 68 (2014) 508–523

800

513

60 Urban Population

700

50

Urbanization Rate 600

(Millions)

400

30

300

(%)

40

500

20

200 10

0

1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

100

0

Fig. 7. Urbanization in China, 1980–2011. Table 3 Annual speed settings for urbanization in China in alternative scenarios. Period

High speed (%)

Baseline (%)

Low speed (%)

2011–2015 2016–2020 2012–2025 2026–2030 2031–2035 2036–2040 2041–2045 2046–2050

1.2 1.2 1.0 1.0 0.8 0.8 0.8 0.8

1.0 1.0 0.8 0.8 0.6 0.6 0.6 0.6

0.8 0.8 0.6 0.6 0.4 0.4 0.4 0.4

4.5.2. Projection of energy intensity for China, 2010–2050 According to the 12th FYP (2011–2015) (State Council, 2013), 16% reduction in energy intensity is estimated to happen in 2015 based on the 2010 level. Despite that the 12th FYP target is 20% lower than the one during the 11th FYP period; it is still regarded as a challenging target for China (Yuan et al., 2013). Nevertheless, greater reduction can be expected from energy efficiency applications. In This study provides two scenarios: the baseline scenario is largely an extension of China's past trend, and an acceleration scenario assumes that additional market-oriented mechanisms will be implemented in the future (Table 6). 4.6. Resident income

Table 4 Urbanization scenarios for China, 2010–2050. Period

High speed (%)

Baseline (%)

Low speed (%)

2010 2015 2020 2025 2030 2035 2040 2045 2050

49.95 55.95 61.95 66.95 71.95 75.95 79.95 83.95 87.95

49.95 54.95 59.95 63.95 67.95 70.95 73.95 76.95 79.95

49.95 53.95 57.95 60.95 63.95 65.95 67.95 69.95 71.95

is expected to slow down to 0.8–1.0 percent annually in the coming decade. The speed is expected to further slow down after 2020. Three scenarios for urbanization in 2050 are provided in Tables 3 and 4. The baseline scenario is that urbanization will reach 60% in 2020, 68% in 2030 and 80% in 2050.

4.5. Energy intensity 4.5.1. Change of energy intensity in China Secondary industry holds the dominant position in primary energy consumption, roughly accounting for 70% of it (Fig. 8). Energy efficiency was introduced in China in 1990s, which has a great potential due to technology improvement and optimized industrial structure. Energy intensity of the economy (in 2005 constant price) was 1.79 tce in 1994 and then decreased to 1.03 tce in 2010, which shows a declining ratio of this decreasing trend in 2010s comparing with that in the 1990s. However, with the transition of industrial structure and technological progress, there is great potential to improve China's energy efficiency level (Table 5).

4.6.1. Change of resident income in China, 1980–2011 Urban-rural dual structure in China is an issue that must be properly resolved in China's development process. Urban and rural residential income continues to increase along with economic growth (Fig. 9) (NSBC, 2012). However, the urban-rural income gap has continuously been enlarged from 3.07:1 in 1980 to 3.23:1 in 2012 (Fig. 10). According to International Labor Organization (ILO, 2012), the urban-rural income gap in most countries is around 1.5:1. Here we define the ratio of change in per capita residential income related to change per capita GDP as income elasticity coefficient. Table 7 reports annual average coefficients for rural and urban residents at five years interval. 4.6.2. Projection of resident income The Chinese Government has committed to improve income distribution. In order to realize this commitment, two synchronisms, viz. synchronism between economic and resident income growth, and synchronism between labor productivity and salary growth are proposed as the guiding principle for income distribution policy. After entering upper middle income group, China will be confronted with a new challenge of entering into high income group. Based on the present growth rate, it would take 11–15 years to finish the transition. It took Japan 12 years to finish the transaction, took 11 years for Singapore, Hong Kong SAR 11 years, and South Korea 7 years to finish the stride (WDI, 2012). Their experience also reveals that income per capita generally grows faster than per capita GDP around the critical point; and the income elasticity coefficient ranges between 1 and 1.2. Afterwards, the growth of income per capita will slow down and then the coefficient will range between 0.7 and 1. For the rural residential income elasticity, we assume the growth of rural income would be faster than that of urban income.

514

J. Yuan et al. / Energy Policy 68 (2014) 508–523

100 90 80 70

(%)

60 50 40 30 20 10 0 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Primary Industry Energy Consumption

Secondary Industry Energy Consumption

Tertiary Industry Energy Consumption Fig. 8. Change in share of industry energy consumption. Note: In China statistical system changed in 1994 and it is difficult to distinguish energy consumption for primary, secondary and tertiary industries during 1980–1993. Therefore the data covers 1994–2010 period in this figure.

Table 5 Change of industrial and GDP energy intensity in China. Period

Annual decrease rate of primary industry (PI) (%)

Annual decrease rate of secondary industry (SI) (%)

Annual decrease rate of tertiary industry (TI) (%)

Decrease of GDP energy intensity (%)

2000/1995 2005/2000 2010/2005

8.62 7.98 3.02

7.19 0.07 4.86

4.11 0.13 4.39

26.67 18.15 19.06

Table 6 Projection of industrial and GDP energy intensity improvement in China, 2010–2050. Annual decrease rate of SI

Annual decrease rate of TI

Baseline (%)

Acceleration (%)

Baseline (%)

Acceleration (%)

Baseline (%)

Acceleration (%)

0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00

5.30 5.00 4.50 4.00 3.70 3.70 3.40 3.40

5.50 5.20 4.80 4.50 4.20 4.20 3.90 3.90

4.20 4.20 4.20 4.20 4.20 4.20 4.20 4.20

4.50 4.50 4.50 4.50 4.50 4.50 4.50 4.50

25000

25% 20%

20000

15% 10% 10000 5% 5000

0

0%

Annual Per Capita Disposable Income of Urban Households Annual Per Capita Net Income of Rural Households Growth Rate of Annual Per Capita Disposable Income of Urban Households Growth Rate of Annual Per Capita Net Income of Rural Households Fig. 9. Change in resident income in China, 1980–2012.

-5%

(%)

15000

1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

2011–2015 2016–2020 2012–2025 2026–2030 2031–2035 2036–2040 2041–2045 2046–2050

Annual decrease rate of PI

(2005 RMB)

Period

J. Yuan et al. / Energy Policy 68 (2014) 508–523

515

4.00 Urban-rural Income Gap 3.50 3.00 2.50 2.00 1.50 1.00 0.50

1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

0.00

Fig. 10. Change in urban-rural income gap in China, 1980–2012.

Table 7 Annual average income elasticity coefficients in China, 1980–2010. Period

Urban disposable income

Rural net income

1980–1985 1985–1990 1990–1995 1995–2000 2000–2005 2005–2010

0.44 1.08 0.75 0.78 1.08 0.92

1.98 0.35 0.39 0.61 0.58 0.85

Thus, we compile the baseline scenario of income elasticity coefficients and assume that the coefficient will keep at 1 or above near the transition points, which would be decreasing in the middle of a growth stage. Two alternative scenarios are also provided for comparative analysis (Table 8). 4.7. Energy consumption 4.7.1. Change in China's energy consumption, 1980–2011 After the reform and opening-up, China made huge progress in energy efficiency in 1980s and 1990s. From 1980 to 2000, China managed to fuel fourfold GDP growth with double energy consumption growth (Fig. 11) (NSBC, 2012). Coal consumption shares 70% of China's primary energy consumption. (Fig. 12). Mass consumption of coal is the main reasons of China's low energy efficiency. Economic growth and energy-environmentecology constraints will compel the Chinese Government to pay more attention to energy diversity in the future. 4.7.2. Projection of primary energy mix for China, 2010–2050 We set the baseline scenario of primary energy mix based on the 15% clean energy target by 2020.energy endowment, new discovery of China's domestic gas and oil reserve (including shale gas) and China's recent international purchase agreement on oil and natural gas (Yuan et al., 2012, 2014). To probe the possibility other than the Government's plan, we set another reinforced clean energy path in which clean energy share is as assumed to be two percent higher (17%) in 2020 (Table 9).

household consumption in total primary energy consumption declined from 15.9% in 1980 to 10.6% in 2010 (NSBC, 2012). In OECD countries, household energy consumption normally accounts for 19% of total energy consumption (not including the consumption in private transportation) (IEA, 2013). The underlying reason is that consumption in household grows much slower than that of production. There was a decline of per capita energy consumption in urban household from 1980 to 2001. A key contribution was the substitution of coal by cleaner and more efficient energy products, such as gas and electricity. Coal accounted for 89% of urban household energy consumption in 1986, but dropped to 13% while electric power rose to 50% in 2007. The gap between urban and rural energy consumption was getting smaller (Fig. 13). An urban resident consumed 5.5 times of energy of a rural resident in 1980, while the ratio was 2.6 in 2001. However, the consumption structure of rural residents is different from that of urban consumption: for rural resident, coal accounted for 60.4% of household energy consumption while electric power only accounted for 27.2% and oil for 11.8%. 4.8.2. Projection of household energy consumption in China, 2010–2050 What is China's household energy consumption after 2020? Due to the unavailability of the data for developed countries, we use data of China's relatively developed regions as a reference. Table 10 provides the data of Shanghai's economic growth during 2002–2011. Notice that during the period per capita GDP of Shanghai increased from 4103US$ to 12783US$, which is exactly the transition from upper middle to high income. Energy consumption elasticity was significantly large during the period, ranging between 0.6 and 1.0, and then became smaller when Shanghai finished the transition in 2011. Similar phenomenon can be found for Beijing and Shenzhen. We use these findings to set the elasticity coefficients of urban energy consumption. For rural residents, since their absolute level of energy consumption is still low, we assume that the coefficient would be as large as 1 for a long time and then become gradually small. Furthermore, an Acceleration path is provided for comparative analysis (Table 11).

5. Scenarios 4.8. Household energy consumption 5.1. Scenario setting 4.8.1. Change in household energy consumption in China, 1980–2011 Along with economic and income growth, household energy consumption has grown quickly in China. However, the share of

Given the variables and assumptions, the combination will result in a total number of 144 energy consumption scenarios

516

J. Yuan et al. / Energy Policy 68 (2014) 508–523

Table 8 Scenarios of income elasticity coefficients in China, 2010–2050.

2010-2015 2015–2020 2020–2025 2025–2030 2030–2035 2035–2040 2040–2045 2045–2050

Urban per capita disposable income elasticity coefficient

Rural per capita net income elasticity coefficient

High

Baseline

Low

High

Baseline

Low

1.10 1.00 0.90 0.80 1.10 1.00 0.90 0.80

1.05 0.95 0.85 0.75 1.05 0.95 0.85 0.75

1.00 0.90 0.80 0.70 1.00 0.90 0.80 0.70

1.20 1.10 1.00 0.90 1.20 1.10 1.00 0.90

1.15 1.05 0.95 0.85 1.15 1.05 0.95 0.85

1.10 1.00 0.90 0.80 1.10 1.00 0.90 0.80

18%

4000 Energy Consumption

16%

3500

Mtce

Growth Rate

14%

3000

12%

2500

10% 8%

2000 6% 1500

(%)

Period

4% 2%

1000

0% 500 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

0

-2% -4%

Fig. 11. Energy consumption and growth rate in China, 1980–2011.

100 90 80 70

(%)

60 50 40 30 20

0

1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

10

Coal

Oil

Gas

Non-fossil

Fig. 12. Primary energy consumption structure in China, 1980–2011.

and 432 CO2 emissions scenarios. To summarize these scenarios, we set three main scenarios based on GDP growth and then studied 4 (5 in the baseline GDP growth scenario) key subscenarios within the main scenario. Table 12 provides an overall description of the combination of all energy scenarios and Table 13 provides the description for primary energy mix scenarios. 5.2. Scenario results 5.2.1. Energy consumption According to our methodology and preceding assumptions, scenarios for China's energy consumption are calculated (Table 14).

Under baseline GDP growth scenario, per capita GDP would reach 5200US$ in 2015, then go across the high income threshold and reach 12,600US$ in 2030, and then it is expected to reach 28,300US$ in 2050. Energy consumption will peak in 2040 at 5424 Mtce (SE21). In SE22 the peak would be reduced to 5027 Mtce and appear in 2035 (five years earlier than in SE21), due to enhanced energy efficiency improvement. In SE23, the peak will be reduced by 200 Mtce and appear 5 years earlier (in 2030) due to rapid industrial restructuring. And in SE24 the peak will be 4900 Mtce, which is significantly smaller than that in SE21 because of combined effect of energy efficiency and economic restructuring. The comparison between SE21 and SE25 indicates the difference in household energy consumption because

J. Yuan et al. / Energy Policy 68 (2014) 508–523

517

Table 9 Scenario of primary energy mix for China, 2010–2050. Year

2015 2020 2025 2030 2035 2040 2045 2050

Reinforced clean path

Baseline path

Coal (%)

Oil (%)

Gas (%)

Non-fossil (%)

Coal (%)

Oil (%)

Gas (%)

Non-fossil (%)

63.0 58.0 55.0 53.0 51.0 49.0 47.0 45.0

18.5 18.0 17.5 17.0 16.5 16.0 15.5 15.0

6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0

12.0 17.0 20.0 22.0 24.0 26.0 28.0 30.0

63.6 60.0 57.0 55.0 53.0 51.0 49.0 47.0

18.5 18.0 17.5 17.0 16.5 16.0 15.5 15.0

6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0

11.4 15.0 18.0 20.0 22.0 24.0 26.0 28.0

350

20

300

15 10

250

5 0 150

(%)

(kgce)

200

-5 100

-10 -15

0

-20

1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

50

Annual per Capita Energy Consumption of Urban Households Annual per Capita Energy Consumption of Rural Households Growth Rate of Annual per Capita Energy Consumption of Urban Households Growth Rate of Annual per Capita Energy Consumption of Urban Households Fig. 13. Per capita household energy consumption in China, 1980–2010. Table 10 Per capita household energy consumption in Shanghai, 2002–2011.

Per capita GDP (US$) Urban disposable income, per capita (RMB) Rural net income, per capita (RMB) Urban household energy consumption, per capita (Kg sce) Rural household energy consumption, per capita (Kg sce) Energy consumption elasticity coefficient

2002

2003

2004

2005

2009

2010

2011

4103 13,250 6212 355.13 351.18 –

4650 14,867 6658 382.44 385.35 0.80

5417 16,683 7337 423.87 425.07 0.84

6061 18,645 8342 340.89 477.99 1.06

10,125 28,838 12,324 429.63 642.63 0.85

11,238 31,838 13,746 437.33 671.67 0.53

12,783 36,230 15,644 452.69 702.33 0.33

of the high pace of urbanization and improved income distribution. Although the difference is still small until 2020, it will grow bigger after 2020 and increase the peak by 200 Mtce. In SE21, household energy consumption would share 22% of primary energy in 2050; however, it would account for 26% in SE25. Under high GDP growth scenario, China's per capita GDP of China would reach 5400US$ in 2015, go across the high income threshold point in 2027, reach 13,900US$ in 2030 and then reach 34,300US$ in 2050. If all the variables develop in high or acceleration assumption (in SE11), energy consumption will therefore peak in 2035 at 5522 Mtce, which is nearly 100 Mtce higher than in SE21. In SE12 the peak will move up to 6014 Mtce in 2040, since energy efficiency is improved in slower (baseline) speed. In SE13, the peak will appear five years later than in SE11 and move up by 200 Mtce. In SE14 the peak will be 6283 Mtce and appear in 2045, which is 760 Mtce higher than in SE11. The results clearly indicate that energy efficiency and industrial structure have the most remarkable impact on energy consumption in China.

For the low GDP growth scenario, peak energy consumption will appear in SE31 in 2030 at 4750 Mtce, comparing with SE21, it is 670 Mtce less and appears ten years earlier. An interesting finding is that, when GDP grows with 0.5 percent faster than baseline speed, the increase of peak energy consumption would increase a marginal of 100 Mtce; however, when GDP growth is 0.5 percent slower than that of the baseline scenario, there is a substantial decrease of 670 Mtce (Fig. 14). On the one side, economic growth is desirable but energy demand will outnumber feasible range. On the opposite, energy demand is manageable but GDP growth is politically unacceptable. The only viable solution to this dilemma is to significantly improve energy efficiency and accelerate the restructuring of economic output.

5.2.2. CO2 emissions Under primary energy mix envisioned by Chinese Government, CO2 emissions will peak at 9470 Mt in 2030 with baseline GDP

518

J. Yuan et al. / Energy Policy 68 (2014) 508–523

it stepped into high income group. It continued to increase to more than 6 tce when its per capita GDP reached 20,000US$. A confident observation at this time is that at least South Korea follows the routine of earlier industrialized countries like the United Kingdom, France and Japan; but it is too early to predict whether it will follow the trend of the United States. For per capita CO2 emissions, similar trend can be observed from Fig. 16. Table 18 provides the projection of China's main indexes based on baseline scenario of energy consumption (SE21) and CO2 emissions (SA21). In 2030 when China steps into high income group, per capita energy consumption will reach 3.8 tce. It would be 4 tce in 2045 and afterwards it would decline to 3.95 tce. For per capita CO2 emissions, it would reach 6.8 t in 2030 and stay at that peak for 5 years. For GDP energy intensity and CO2 intensity, continuous and consistent improvement can be observed during the whole scenario period. It is unfair to compare China with US and UK. Hence, NICs will be more suitable for a comparative study for China. A convenient and appropriate option is South Korea. South Korea finished the stride from upper middle income to high income during 1977– 1992. Figs. 17–20 provide the comparisons between South Korea and China, in term of per capita energy consumption, per capita CO2 emissions, energy intensity and CO2 intensity. Figures in the left (A) represent the stage from upper middle to high income, and figures in the right (B) represent the stage of high income. The comparison clearly indicates that under the projected scenarios China could possibly choose a different path. Energy consumption and CO2 emissions will grow slower during upper middle income stage, and will come to a peak without much growth soon after income reaches the level of high income group. The underlying force to shape the new path is continuous and remarkable improvement in energy efficiency and prompt decarbonization of primary energy supply. Thus, China could explore a new pattern of energy consumption and CO2 emissions during income growth.

growth and energy efficiency improvement. The peak could be reduced to 8800 Mt if improved energy efficiency and faster industrial restructuring was realized. Under the high-speed GDP growth scenario, the peak will appear in 2035 at 10,660 Mt. According to this energy path, in 2020, CO2 emissions will reach 8667 Mt, which matches with the IEA 450ppm scenario for China's 8600 Mt target. Under a strengthened clean energy path, the emissions peak under high GDP growth scenario (SB11) could even be smaller than that under baseline GDP growth scenario and the official plan of primary energy mix (SA21). The implication is that clean energy development can provide higher potential for China to maintain high economic growth. After 2030, the implementation of CCS will effectively reshape the path of CO2 emissions off the peak and render the reduction more sharply (Tables 15–17).

5.3. Discussion of the results Shall China simply replace experiences from industrialized countries and follow their traditional routine of energy consumption and CO2 emissions? Fig. 15 provides a comparison of per capita energy consumption vs. per capita GDP for China and other countries and Fig. 16 demonstrates the per capita CO2 emissions (IEA, 2009b). For per capita energy consumption, the OECD average stays at roughly 6 tce since their per capita GDP reaches 20,000US$. For the United States and Canada, per capita consumption stabilizes between 10 and 12 tce with limited increase, while for Japan, the United Kingdom and France, their ranges were between 4 and 6 tce. For South Korea, a newly industrialized country (NIC), per capita energy consumption reached 4 tce when Table 11 Projection of household energy consumption elasticity in China, 2010–2050. Period

2010–2015 2015–2020 2020–2025 2025–2030 2030–2035 2035–2040 2040–2045 2045–2050

Urban

Rural

Baseline

Acceleration

Baseline

Acceleration

0.25 0.65 0.65 0.30 0.40 0.40 0.30 0.30

0.20 0.60 0.60 0.25 0.35 0.35 0.25 0.25

1.05 0.95 0.85 0.75 0.65 0.65 0.60 0.55

1.00 0.90 0.80 0.70 0.60 0.60 0.55 0.50

Table 13 The illustration of scenario setting for primary energy mix. Scenario

Description

Baseline (SA) Strengthened (SB) Strengthenedþ CCS (SC)

Primary energy mix changes in baseline path. Primary energy mix changes in reinforced path. In addition to SA, capture 1 percent point CO2 (120–150 Mt) from coal consumption by CCS during 2030–2040 and 100–120 Mt during 2040–2050.

Table 12 The overall description on scenario setting. Scenario

SE11

SE12

SE13

SE14

High Baseline Low

√

√

√

√

Industrial structure

Acceleration Baseline

√

√

Urbanization

High Baseline Low

√

√

Energy efficiency

Acceleration Baseline

√

Per capita income

High Baseline Low

√

High Baseline

√

GDP

Household energy

√

√

√

√

SE21

SE22

SE23

SE24

SE25

√

√

√

√

√

√

√

√

√ √

√

√

SE33

√

√

√

√

√

√

√

√

√

√

SE34

√

√ √

√

√

√ √

√

√

√ √

√

√ √

√

√ √

√ √

√

SE32

√ √

√

√

SE31

√

√

√

√

√

√

√

√

√

√

√

√

√ √

√

√

√

J. Yuan et al. / Energy Policy 68 (2014) 508–523

519

Table 14 Energy consumption scenarios of China, 2010–2050 (Mtce). Year

SE11

SE12

SE13

SE14

SE21

SE22

SE23

SE24

SE25

SE31

SE32

SE33

SE34

2010 2015 2020 2025 2030 2035 2040 2045 2050

3250 4012 4533 5098 5407 5522 5516 5372 5110

3250 4059 4636 5287 5711 5931 6014 5939 5716

3250 4048 4612 5186 5504 5679 5731 5649 5453

3250 4095 4718 5380 5821 6115 6274 6283 6153

3250 3990 4507 5013 5277 5415 5424 5327 5114

3250 3945 4406 4832 4988 5027 4952 4789 4535

3250 3955 4429 4926 5176 5251 5198 5035 4752

3250 3910 4330 4750 4899 4888 4765 4555 4251

3250 4012 4531 5068 5382 5555 5606 5523 5326

3250 3888 4273 4627 4747 4741 4623 4423 4138

3250 3843 4177 4458 4484 4397 4214 3968 3659

3250 3853 4199 4545 4655 4596 4427 4176 3839

3250 3809 4105 4382 4403 4273 4052 3770 3425

6500

6000

6000

5500

Energy Consumption (Million tce)

Energy Consumption (Million tce)

Note: Figures in shadowed column is the baseline projection and used for international comparative study.

5500 5000 4500 4000 3500

5000 4500 4000 3500 3000

3000 2010

2015

2020

2025

SE11

2030

2035

SE12

SE13

2040

2045

2010

2050

SE14

2015 SE21

2020

2025

SE22

2030

2035

SE23

2040 SE24

2045

2050

SE25

6000

5000

Energy Consumption (Million tce)

Energy Consumption (Million tce)

4800 4600 4400 4200 4000 3800 3600 3400

5500 5000 4500 4000 3500

3200 3000

3000 2010

2015

2020

2025

SE31

2030

2035

SE32

SE33

2040

2045

2050

2010

2015

SE34

2020

2025 SE11

2030 SE21

2035

2040

2045

2050

SE31

Fig. 14. Comparison of various energy consumption scenario results.

Table 15 SA CO2 emissions scenarios of China, 2010–2050 (Mt). Year

SA11

SA12

SA13

SA14

SA21

SA22

SA23

SA24

SA25

SA31

SA32

SA33

SA34

2010 2015 2020 2025 2030 2035 2040 2045 2050

7669 8082 8717 9414 9706 9627 9332 8811 8117

7669 8175 8915 9762 10,250 10,341 10,174 9740 9079

7669 8154 8870 9577 9880 9900 9696 9264 8661

7669 8248 9073 9935 10,449 10,660 10,614 10,305 9774

7669 8038 8667 9258 9472 9440 9175 8736 8123

7669 7945 8473 8923 8953 8764 8378 7854 7203

7669 7966 8516 9096 9290 9155 8793 8258 7548

7669 7875 8327 8772 8795 8521 8061 7470 6753

7669 8081 8713 9359 9660 9684 9484 9057 8460

7669 7831 8218 8544 8520 8266 7820 7254 6572

7669 7741 8033 8233 8049 7665 7129 6508 5812

7669 7761 8074 8394 8355 8013 7488 6849 6097

7669 7672 7893 8091 7904 7449 6854 6182 5440

Note: Figures in shadowed column is the baseline projection and used for international comparative study.

6. Policy implication and conclusion In the next 40 years, China will experience rapid industrialization and urbanization processes. An inevitable companion is the growth of energy consumption. In this paper, we construct various

scenarios of energy demand and CO2 emissions by considering the most important driving factors and by simulating various socialeconomic policies. This research leads to three findings. Firstly, characteristic of China’s energy consumption and CO2 emissions at each phase is different. Before 2020, due to the

520

J. Yuan et al. / Energy Policy 68 (2014) 508–523

Table 16 SB CO2 emissions scenarios of China, 2010–2050 (Mt). Year

SB11

SB12

SB13

SB14

SB21

SB22

SB23

SB24

SB25

SB31

SB32

SB33

SB34

2010 2015 2020 2025 2030 2035 2040 2045 2050

7669 8022 8493 9163 9439 9355 9059 8546 7865

7669 8115 8686 9501 9968 10,048 9877 9447 8797

7669 8094 8642 9321 9608 9620 9413 8985 8392

7669 8187 8840 9670 10,162 10,358 10,304 9994 9470

7669 7978 8444 9010 9211 9172 8907 8473 7870

7669 7887 8255 8685 8707 8515 8133 7618 6979

7669 7907 8298 8853 9035 8896 8536 8009 7313

7669 7817 8113 8537 8552 8280 7826 7245 6543

7669 8021 8490 9109 9395 9409 9207 8785 8197

7669 7773 8007 8316 8286 8031 7592 7036 6368

7669 7684 7826 8012 7827 7448 6920 6312 5631

7669 7704 7867 8169 8125 7786 7270 6643 5908

7669 7616 7690 7875 7686 7238 6654 5996 5271

Table 17 SC CO2 emissions scenarios of China, 2010–2050 (Mt). SC11

SC12

SC13

SC14

SC21

SC22

SC23

SC24

SC25

SC31

SC32

SC33

SC34

2010 2015 2020 2025 2030 2035 2040 2045 2050

7669 8022 8493 9163 9439 9261 8968 8460 7786

7669 8115 8686 9501 9968 9947 9778 9353 8709

7669 8094 8642 9321 9608 9524 9319 8895 8308

7669 8187 8840 9670 10,162 10,255 10,201 9895 9375

7669 7978 8444 9010 9211 9080 8818 8388 7791

7669 7887 8255 8685 8707 8430 8052 7542 6909

7669 7907 8298 8853 9035 8807 8451 7929 7240

7669 7817 8113 8537 8552 8197 7748 7173 6477

7669 8021 8490 9109 9395 9315 9115 8697 8115

7669 7773 8007 8316 8286 7951 7516 6965 6304

7669 7684 7826 8012 7827 7373 6851 6248 5575

7669 7704 7867 8169 8125 7708 7197 6577 5849

7669 7616 7690 7875 7686 7166 6587 5936 5218

Energy Consumption per capita(tce/person)

Year

14 12 10 8 6 4 2 0

0

10000

20000 30000 40000 GDP per capita (2012 US$)

50000

60000

World

OECD

Canada

United States

Japan

Korea

France

United Kingdom

China

Fig. 15. Per capita energy consumption vs. GDP per capita in the world, 1971–2010.Note: in IEA the US$ is in 2005-year constant price. For consistence of comparison, we recalculate the data according to the change in exchange rate between RMB and US$ from 2005 to 2012.

accelerated process of industrialization and growth in residential energy consumption, China's energy consumption and CO2 emissions will stay at high growth rate even with remarkable improvement in energy efficiency. China is projected to step in the later stage of industrialization from 2020 to 2030 and the process of urbanization shall be reduced. Due to the continuous improvement of energy efficiency and rapid expansion of non-fossil energy, the growth of energy consumption and CO2 emissions would be reduced. During this period, non-fossil energy will begin to substantially reshape the China's primary energy mix. Beyond 2030, China will step into high income group and energy demand will begin to saturate and eventually reach its plateau. With increasing deployment of renewable energy and implementation of CCS, CO2 intensity will be improved in a faster pace than that of GDP energy intensity.

The second finding is that energy consumption as well as CO2 emissions will reach a peak in the next 10 years. Under the planned GDP growth path, energy consumption will peak at about 5300 Mtce in 2040. Combining with the clean energy path envisioned by the Government, CO2 emissions is estimated to peak at 9400 Mt in 2030–2035. A reinforced clean path in combination with the CCS deployment will reduce the peak by 200–300 Mt. Meanwhile, our scenario offers an optimistic prospective that with the consistent improvement of energy efficiency and substantial development of clean energy, it is highly possible for China to break the reliance on energy and decouple CO2 emissions from its economic growth during the transition phase from upper middle income to high income. The third finding, as well as the most important policy implication of the study is that, continuous energy efficiency

J. Yuan et al. / Energy Policy 68 (2014) 508–523

521

CO2 Emissions per capita(tonnes)

25 20 15 10 5 0

0

10000

20000

30000

40000

50000

60000

GDP per capita (2012 US$) World

OECD

Canada

United States

Japan

Korea

France

United Kingdom

China

Fig. 16. Per capita CO2 emissions vs. GDP per capita in the world, 1971–2010.

Table 18 China's main indexes under baseline energy consumption and CO2 emissions growth, 2010–2050. Per capita energy consumption (tce)

Per capita CO2 emissions (tonnes)

GDP energy intensity (tce/thousand US$)

GDP CO2 intensity (ton/thousand US$)

2010 2015 2020 2025 2030 2035 2040 2045 2050

3725 5235 7247 9876 12,625 15,864 19,594 23,781 28,338

2.42 2.91 3.25 3.59 3.78 3.92 3.98 3.99 3.95

5.71 5.86 6.24 6.63 6.80 6.83 6.74 6.56 6.27

0.680 0.550 0.442 0.362 0.304 0.247 0.203 0.168 0.139

1.608 1.120 0.861 0.671 0.538 0.430 0.344 0.275 0.221

Energy Consumption per capita(tce/person)

Per capita GDP (2012 US$)

Energy Consumption per capita(tce/person)

Year

4.0 3.5 3.0 2.5 2.0 1.5 1.0

Korea

0.5 China 0.0 0

2000

4000

6000

8000

10000

12000

14000

8.0 7.0 6.0 5.0 4.0 3.0 2.0

Korea

1.0 China 0.0 12000

14000

16000

GDP per capita (US$)

18000

20000

22000

24000

26000

28000

30000

GDP per capita (US$)

8.0

14.0

7.0

12.0

CO2 Emissions per capita(tonnes)

CO2 Emissions per capita(tonnes)

Fig. 17. Comparison of per capita energy consumption vs. per capita GDP in China and South Korea.

6.0 5.0 4.0 3.0 2.0

Korea

1.0 China 0.0 0

2000

4000

6000

8000

GDP per capita (US$)

10000

12000

14000

10.0 8.0 6.0 4.0 Korea 2.0 China 0.0 12000

17000

22000 GDP per capita (US$)

Fig. 18. Comparison of per capita CO2 emissions vs. per capita GDP in China and South Korea.

27000

32000

522

J. Yuan et al. / Energy Policy 68 (2014) 508–523

0.8

GDP Energy Intensity(kgce US$)

GDP Energy Intensity(kgce US$)

0.35 0.7 0.6 0.5 0.4 0.3 0.2 Korea

0.1 0

China 0

2000

4000

6000

8000

10000

12000

0.30 0.25 0.20 0.15 0.10 Korea 0.05 0.00 12000

14000

China 14000

16000

GDP per capita (US$)

18000

20000

22000

24000

26000

28000

30000

GDP per capita (US$)

Fig. 19. Comparison of GDP energy intensity vs. per capita GDP in China and South Korea.

0.6

1.80

GDP CO2 Intensity(kgce/ US$)

GDP CO2 Intensity(kgce/ US$)

1.60 1.40 1.20 1.00 0.80 0.60 0.40 Korea China

0.20 0.00

0

2000

4000

6000

8000

10000

12000

14000

0.5 0.4 0.3 0.2 0.1 0 12000

14000

16000

GDP per capita (US$)

18000

20000

22000

24000

26000

28000

30000

GDP per capita (US$)

Fig. 20. Comparison of GDP CO2 intensity vs. per capita GDP in China and South Korea.

improvement, radical industrial restructuring and rational urbanization are three driving forces of the new pathway. Energy intensity would be decreased at 4% per year in the coming 30–40 years. Enhanced R&D, industrial and fiscal policies together with improved energy efficiency and environmental standards, could boost innovation in energy efficiency and promote the development of clean technologies. Structural energy conservation will become increasingly important after China steps into the later stage of industrialization. Since the residential sector will consume substantial amount of primary energy, it is also important to pay attention to the urbanization process and associated energy conservation potentials. Optimizing the transportation systems and building standards would be a long-term challenge for China to effectively solve the energy bottleneck. The contribution of the paper is twofold. Firstly, a methodology for energy and CO2 emissions scenario analysis is proposed. It can explicitly model the impact of GDP growth, industrial structure, energy intensity, population, urbanization and income distribution on China’s energy consumption. This methodology is adapted from Kaya identity and practically suitable for China. Secondly, various scenarios are constructed to explore China's future trajectories on energy consumption and CO2 emissions. China could restructure its primary energy mix after realizing energy efficiency standardization, rational urbanization and clean energy development. But it is worth noting that the coming of peak energy in China in around 2040 in baseline GDP growth is based on strong assumption of 4% annual GDP energy intensity reduction for consecutive 30 years. Therefore, the study in the paper just reveals the possibility, and cannot be mistakenly understood that peak energy will certainly appear in 2040. The idea presented in the paper can also be a source of references for developing countries that are similar to China.

Acknowledgments The work reported in the paper is funded by the Ministry of Education of China (10YJC790360), the National Science Foundation of China (71173075), the Beijing Higher Education Young Elite Teacher Project and the Fundamental Research Funds for the Central Universities. The usual caveats apply. References Aldy, J.E., 2006. Per capita carbon dioxide emissions: convergence or divergence? Environ. Resour. Econ. 33, 533–555. Barassi, M.R., Cole, M.A., Elliott, R.J.B., 2008. Stochastic divergence or convergence of per capita carbon dioxide emissions: re-examining the evidence. Environ. Resour. Econ. 40, 121–137. CEACER (2050 China Energy and CO2 Emissions Research Group), 2009. 2050 China Energy and CO2 Emissions Report. Science Press, Beijing CAE (Chinese Academy of Engineering), 2011. China Energy Medium and Long-term (2030, 2050) Development Strategy Research. Science Press, Beijing Grossman, G.M., Krueger, A.B., 1995. Economic growth and the environment. Q. J. Econ. 110, 353–377. Grossman, G.M., Krueger, A.B., 1991. Environmental impacts of a North American free trade agreement, National Bureau of Economic Research Working Paper 3914. NBER, Cambridge, MA Hu J., 2012. Report on the 18th National Congress of CPC. 〈http://www.xj.xinhuanet. com/2012-11/19/c_113722546.htm〉 (accessed 2012]. IBRD, 1992. World development report 1992, Development and the Environment. Oxford University Press, New York IEA (International Energy Agency), 2009a. World Energy Outlook 2009. OECD Publishing, Paris IEA, 2009b. CO2 Emissions from Fuel Combustion (2009 ed.). IEA, Paris IEA, 2013. Energy Balance for OECD countries. IEA, Paris IPCC, 2007. In: Metz, B., Davidson, O.R., Bosch, P.R., Dave, R., Meyer, L.A. (Eds.), Climate Change 2007: Mitigation, Contribution of Working Group III to the Fourth Assessment Report of the IPCC. Cambridge University Press, Cambridge and New York, USA. Cambridge, United Kingdom and New York, NY, USA. ILO (International Labor Organization), 2012. Household income and expenditure. 〈http://laborsta.ilo.org/〉.

J. Yuan et al. / Energy Policy 68 (2014) 508–523

Iwata, H., Okada, K., Samreth, S., 2011. A note on the environmental Kuznets curve for CO2: a pooled mean group approach. Appl. Energy 88, 1986–1996. Jakob, M., Haller, M., Marschinski, R., 2012. Will history repeat itself? Economic convergence and convergence in energy use patterns. Energy Econ. 34, 95–104. Jalil, A., Mahmud, S.F., 2009. Environment Kuznets curve for CO2 emissions: a cointegration analysis for China. Energy Policy 12, 5167–5172. Jaunky, V.C., 2011. The CO2 emissions–income nexus: evidence from rich countries. Energy Policy 39 (3), 1228–1240. Jiang, K., Hu, X., Zhuang, X., Liu, Q., 2009. China's low-carbon scenarios and roadmap for 2050. Sino-Glob. Energy 6 (6), 21–26. Jobert, T., Karanfil, F., Tykhonenko, A., 2010. Convergence of per capita carbon dioxide emissions in the EU: legend or reality? Energy Econ. 32 (6), 1364–1373. Kaya, Y., Yokobori, K., 1997. Environment, Energy and Economy: Strategies for Sustainability. United Nation University Press, Hongkong Lee, C.C., Chang, C.P., 2008. New evidence on the convergence of per capita carbon dioxide emissions from panel seemingly unrelated regressions augmented Dickey–Fuller tests. Energy 33, 1468–1475. Lin, B., Li, X., 2013. Global convergence in per capita CO2 emissions. Renew. Sustain. Energy Rev. 24, 357–363. Lin, B., Jiang, J., 2009. China' CO2 emission forecast by Kuznets curve and factor analysis. Manag. World 4 (4), 27–36. Luzzati, T., Orsini, M., 2009. Investigating the energy-environmental Kuznets curve. Energy 34 (3), 291–300. McKinsey and Company, 2009. China' Green Revolution: Prioritizing Technologies to Achieve Energy and Environmental Sustainability. McKinsey and Company, Shanghai Meng, M., Payne, J.E., Lee, J., 2013. Convergence in per capita energy use among OECD countries. Energy Econ. 36, 536–545. NSBC (National Statistical Bureau of China), 2001. The fifth population census (2000). 〈http://www.stats.gov.cn/tjsj/ndsj/renkoupucha/2000pucha/pucha. htm〉 (accessed 2012). NSBC, 2011. The sixth population census (2010). 〈http://www.stats.gov.cn/tjsj/pcsj/ rkpc/6rp/indexch.htm〉 (accessed 2012). NSBC, 2012. China statistical book 2012. 〈http://www.stats.gov.cn/tjsj/ndsj/2012/ indexch.htm〉 (accessed 2013). NDRC, 2013. State Council Report on the Urbanization Work. Panopoulou, E., Pantelidis, T., 2009. Club convergence in carbon dioxide emissions. Environ. Resour. Econ. 44, 47–70. Raupach M.R., Marland G., Ciais P., et al., 2007. Global and regional drivers of accelerating CO2 emissions. In: Proceedings of the National Academy of Sciences of the United States of America 104, pp. 10288–10293. Richmond, K.A., Kaufmann, R.K., 2006. Is there a inflection point in the relationship between income and energy use and/or carbon emissions? Ecol. Econ. 56 (2), 176–189. Romero-Ávila, D., 2008. Convergence in carbon dioxide emissions among industrialized countries revisited. Energy Econ. 30, 2265–2282. Rout, U.K., Voβ, A., Singh, A., et al., 2011. Energy and emissions forecast of China over a long-time horizon. Energy 36, 1–11.

523

Selden, T.M., Song, D., 1994. Environmental quality and development: is there a Kuznets curve for air pollution emissions? J. Environ. Econ. Manag. 27, 147–162. Stern, D.I., 2004. The rise and fall of the environmental Kuznets curve. World Dev. 32 (8), 1419–1439. State Council of China, 2011. The 12th Five-year-plan for China's Population Development. State Council of China, 2013. Energy development plan for the 12th FYP. 〈http:// www.gov.cn/zwgk/2013-01/23/content_2318554.htm〉 (accessed 2013). Stegman A., 2005. Convergence in carbon emissions per capita. CAMA Working Papers. Strazicich, M., List, J., 2003. Are CO2 emissions levels converging among industrial countries? Eur. Assoc. Environ. Resour. Econ. 24, 263–271. Steckel, J.C., Jakob, M., Marschinski, R., Luderer, G., 2011. From carbonization to decarbonization? Past trends and future scenarios for China's CO2 emissions. Energy Policy 39 (6), 3443–3455. Sun, J.W., 1999. The nature of CO2 emission Kuznets curve. Energy Policy 27, 691–694. UN (United Nations), 2012. The Population Division of the Department of Economic and Social Affair. World population prospects: the 2012 revision. 〈http://esa.un. org/unpd/wpp/Excel-Data/fertility.htm〉 (accessed 2013). Van, P.N., 2005. Distribution dynamics of CO2 emissions. Environ. Resour. Econ. 32, 495–508. Vuuren van, Zhou, D., et al., 2003. Energy and emission scenarios for China in the 21st century–exploration of baseline development and mitigation options. Energy Policy 31 (4), 369–387. Westerlund, J., Basher, S.A., 2008. Testing the convergence in carbon dioxide emissions using a century of panel data. Environ. Resour. Econ. 40, 109–120. Wang T., Watson J., 2009. China's Energy Transition: Pathways for Low Carbon Development. 〈http://www.sussex.ac.uk/sussexenergygroup/documents/chi na_report_forweb.pdf〉. Brighton, University of Sussex Energy Group. Wang, T., Watson, J., 2010. Scenario analysis of China's emissions pathways in the 21st century for low carbon transition. Energy Policy 38 (7), 3537–3546. WBG (World Bank Group), 2012a. How we classify countries. 〈http://data.world bank.org/about/country-classifications〉. WBG (World Bank Group), 2012b. World development report. WDI (World Development Indicators), 2012. Economic Policy & External Debt. World Bank Database. Yuan, J., Kang, J., Yu, C., Hu, Z., 2011. Energy conservation and emissions reduction in China–progress and prospective. Renew. Sustain. Energy Rev. 15 (9), 4334–4347. Yuan, J., Hou, Y., Xu, M., 2012. China's 2020 carbon intensity target: consistency, implementations, and policy implications. Renew. Sustain. Energy Rev. 16 (7), 4970–4981. Yuan, J., et al., 2014. China's 2020 clean energy target: consistency, pathways and policy implications. Energy Policy 65, 692–700. Zhou, N., Fridley, D., Khanna, N.Z., et al., 2013. China's energy and emissions outlook to 2050: perspectives from bottom-up energy end-use model. Energy Policy 53, 51–62.