sustainability

Article The Effects of an Energy Use Paradigm Shift on Carbon Emissions: A Simulation Study

Yuzhe Wu 1, Jiaojiao Luo 1,2,* ID , Liyin Shen 3 and Martin Skitmore 4

1 Department of Land Management, Zhejiang University, Hangzhou 310058, China; [email protected] 2 Center for International Earth Science Information Network (CIESIN), Columbia University, Palisades, NY 10964, USA 3 School of Construction Management and Real Estate, Chongqing University, Chongqing 400044, China; [email protected] 4 School of Civil Engineering and Built Environment, Queensland University of Technology (QUT), Brisbane 4000, Australia; [email protected] * Correspondence: [email protected]

 Received: 13 April 2018; Accepted: 17 May 2018; Published: 19 May 2018 

Abstract: emissions in developing countries are closely tied to their economy and play a crucial role in the world’s future emissions. In this paper, we put forward an alternative energy use paradigm shift of low-carbon emissions from operational, governance, institutional, and cultural viewpoints (OGIC). An factor is introduced into the Kaya identity, and three simulations are conducted to forecast the and to explore the effects of the energy use paradigm shift policy. The simulation results show that, in the context of the energy use paradigm shift, the years 2015 and 2024 are the two inflection points that separate the carbon footprint into three periods of extensive consumption (2000–2015), early energy transition (2016–2023), and late energy transition (2024–2030). Overall, the peak carbon emission value is forecasted to appear during the third stage. The findings are expected to demonstrate the effects of the energy use paradigm shift on carbon emissions and assist policy makers formulate a scientific policy framework for low carbon development.

Keywords: low carbon emissions; energy use paradigm shift; Kaya identity; OGIC analysis

1. Introduction Carbon emissions, also known as a key component of emissions, are considered to be the main cause of global and seriously threaten the of society [1–3]. Industrialization is often accused of having such an adverse effect as releasing large amounts of carbon dioxide and generating serious global environmental , which seriously affect normal social production activities and human life. At present, the carbon emissions caused by the consumption of natural resources and energy has reached an extent not experienced in the past. The climate change that takes place due to this concentration of carbon emissions is largely irreversible [4,5]. Carbon emissions are not only a pollution that triggers the greenhouse effect and brings about an ecological crisis, but also a development right that is related to a country’s industrialization and urbanization [6]. Rapid shift to a green growth paradigm turns to be a vital determinant of the long-term economic growth [7]. As a result, many international treaties and national policies have been promulgated to deal with carbon emissions and slow climate change. The earliest international treaties date back to the 1992 United Nations Framework Convention on Climate Change. This treated developed and developing countries differently in terms of their obligations and procedures to carry out the commitments involved. However, it was not legally

Sustainability 2018, 10, 1639; doi:10.3390/su10051639 www.mdpi.com/journal/sustainability Sustainability 2018, 10, 1639 2 of 18 binding, as it imposed neither specific obligations nor implementation mechanisms on the individual contracting parties. The Kyoto Protocol emerged soon after in 1997, requiring more than 30 developed and economic transitional countries to reduce their greenhouse gases by 5.2% by 2010 relative to 1990. The first commitment period of the Protocol expired in 2012, but it failed to some extent in the years before the expiration. Most recently, in December 2015, 195 countries met in Paris to adopt a new global agreement on climate change—the Paris Agreement, which was the first sustained, effective, and legally binding agreement in history to be adopted by all parties. Since then, global climate governance has moved to a new stage. At the same time, numerous academic studies of carbon emissions were also conducted. Some focus on demonstrating whether the relationship between environmental degradation and income conforms to environmental Kuznets curve (EKC) theory [8–12]. Specifically, the EKC theory hypothesizes that there is an inverted U-shape relationship between environment change and economic development. Ibrahim and Law [13] examined the mitigating effect of social capital on EKC for carbon emissions using data for 69 developed and developing countries and confirmed the existence of EKC. By applying unit root tests and the VECM Granger causality approach, Shahbaz, et al. [14] also validated the presence of EKC with economic data for Turkey. Jalil and Mahmud [9] examined the long-run relationship between carbon emissions and income in China, finding a quadratic relationship between income and CO2 emissions over the sample period, supporting an EKC relationship; whereas, Stern, et al. [15] commented that most EKC estimates were dependent on the incorrect assumption that world per capita income was normally distributed. There has been insufficient evidence to date to show that countries follow a common inverted U-shaped pathway as their income rises [16]. Other research has specialized in the prediction of future carbon emissions. Most adopt decomposition methods. Zhang and Zhou [17], for instance, established a cointegration model and state space model to support the long-term equilibrium relationship between urbanization level and carbon emissions; while Lin and Liu [18] applied this relationship to further forecast China’s carbon emissions growth through the Kaya identity. Wu, Shen, Zhang, Skitmore and Lu [1] present U-Kaya, a modified version of the Kaya Identity formula, in forecasting China’s likely carbon emissions by 2020. The majority of these studies focus on the relationship between income and carbon emissions or emission forecasts. Comparative studies of China’s energy policy and carbon footprint are much fewer. However, such empirical studies are important, as they can provide a scientific direction for policy makers to obtain social sustainable development. China passed the United States in 2011 as the world’s largest CO2 emitter. In the face of the large demand for energy, a low-carbon pathway should become the key policy priority in China [19–21]. Urbanization and industrialization often make it difficult to reduce carbon emissions effectively without scientific policy guidance. In developed cities, a higher density brought about by agglomeration has placed an excessive burden on the local environment’s carbon absorption capacity [22,23].Therefore, in this paper, we extend the Kaya identity with a new factor of “urbanization rate”, an element closely related to energy consumption emissions in developing countries, to make a carbon footprint forecast with the guidance of an established energy use paradigm shift. Data are collected and Monte Carlo simulation is adopted to forecast the country’s carbon emissions with and without the guidance of an established energy policy. The remainder of this paper is organized as follows. Section2 presents the research method used in this paper. An alternative energy use paradigm shift of carbon emissions in terms of levels of operation, governance, institution and culture is put forward in Section3, while Section4 describes a case study with three simulations to predict the carbon emissions values in China. The final section contains some concluding remarks.

2. Research Method and Data Three simulations were designed to predict the amount of carbon emissions in 2030 by regression analysis, Kaya Identity and Monte-Carlo Simulation. The first simulation is a simple regression with Sustainability 2018, 10, 1639 3 of 18 carbon emission data from previous years. In the second simulation, the parameter setting is based on the existing energy use pattern. The third simulation is designed to forecast the carbon emissions in the context of the energy use paradigm shift.

2.1. Kaya Identity and Model Refinement The Kaya identity was first proposed by Japanese scholar, Yoichi Kaya, in a seminar held by the United Nations Government Climate Change Committee in 1989. In this equation, carbon dioxide emissions are broken down into four human production and life relating factors, namely the energy carbon emissions coefficient (C/E), energy intensity (E/GDP) (gross domestic product), per capita GDP (GDP/P), and population [24]. The mathematical expression is:

C E GDP C = × × × P (1) E GDP P where C represents the amount of carbon dioxide emissions, E is total primary energy consumption, GDP is gross domestic product, and P is the total population. Simple to understand, the C/E refers to the amount of carbon dioxide released from per unit of primary energy consumption, and E/GDP implies how much energy is consumed to realize one unit of GDP. Until the now, Kaya identity has been adopted in a number of carbon emission studies [25–27]. For an industrial society, carbon emissions present a major threat to the sustainability and viability of its economic system. No matter which country or region is involved, it is unlikely that agriculture alone will make much progress in the economy in the face of the wide range of and economic agglomeration [28]. Thus, in order to meet people’s higher standard of living and promote local prosperity, energy-intensive industries, typically a result of industrialization, induce mass carbon emissions and therefore receive the most attention. The energy consumption structure, together with urban and rural residents’ life styles in the different stages of urbanization, will influence carbon emissions too [29,30]. A new factor—“urbanization”—is therefore introduced into the Kaya identity to improve the efficiency of per capita GDP as:

C E GDP GDP C = × × ( u × U + r × (1 − U)) × P (2) E GDP Pu Pr where the new element GDPu/Pu is per capita urban GDP, equal to urban GDP divided by the urban population. Similarly, GDPr/Pr is per capita rural GDP and U is the urbanization rate.

2.2. Monte Carlo Simulation Monte Carlo simulation is a method of studying distribution characteristics by setting up a random process to repeatedly generate a time series and calculate parameter estimates and statistics. It is a widely used technique in the probabilistic analysis of sophisticated systems [31]. Future carbon emissions are hard to forecast because they are a function of inherently unpredictable socio-economic activities [32–34]. Monte Carlo simulation is adopted in the forecasting process in an attempt to accommodate such uncertainty. When the object is probabilistic, a statistic can be calculated based on the sampling distribution generated by the method [35].

2.3.Data The carbon dioxide emission (C) data of 2000–2015 analyzed in this study are obtained from the European Commission (http://edgar.jrc.ec.europe.eu/). The 2000–2015 primary energy consumption (E), GDP, population (P), and urbanization (U) data are obtained from China Statistical Yearbook 2001–2016. Sustainability 2018, 10, 1639 4 of 18

3. Alternative Energy Use Paradigm Shift The present energy use pattern in China is still dominated by the traditional approach. As is well known,Sustainability China 2018 is, 10 in, x an FOR early PEER stage REVIEW of industrialization with an irrational energy structure, and relatively4 of 17 low energy technology equipment and management, which unavoidably leads to a higher energy consumptionenergy consumption per unit GDP per unit than GDPthe world than average the world [36]. average There exists [36]. a There significant exists heterogeneous a significant lowheterogeneous consumption low distribution consumption in distribution the southeast in andthe southeast the northwest, and the while northwest, it is high while in theit is centralhigh in andthe northern central and regions northern [37]. regions Coal and [37 oil]. Coal account and for oil more account than for 80% more of total than energy 80% of consumption total energy (seeconsumption Figure1). (see Figure 1). InIn 2015, 2015, the the total total energy energy consumption consumption of of China China reached reached 4300 4300 million million tons tons coal co equivalental equivalent (Mtce), (Mtce), of whichof which coal consumedcoal consumed 2752 Mtce 2752 and Mtce oil andconsumed oil consumed 774 Mtce, 774 accounting Mtce, accounting for 64% and for 18%, 64% respectively. and 18%, Therespectively. sum of all The clean sum energy, of all such clean as natural energy, gas, such electric as natural power gas, and other electric energy power sources, and other only equalsenergy 774sources, Mtce—a only mere equals 18% 774 of Mtce the total.—a mere Coal 18% is the of leadingthe total. source Coal is of the carbon leading emissions, source of occupying carbon emissions, a much higheroccupying proportion a much (more higher than proportion twice the (more world than average) twice inthe China world while average) natural in gas,China part while and natural parcel ofgas, clean part energy, and parcel is less of than clean a quarterenergy, of is theless world than average.a quarter Inof short,the world actively average. developing In short, renewable actively energydevelop anding newrenewable energy, energy and optimizing and new energy, the energy and consumptionoptimizing the structure energy consumption in China, are structure important in aspectsChina, inare the important control ofaspects greenhouse in the gascontrol emissions of greenhouse in the future. gas emissions in the future.

FigureFigure 1.1.2000–2015 2000–2015 energyenergy consumptionconsumption inin China.China.(Source: (Source: ChinaChina StatisticalStatistical YearbookYearbook 2016). 2016).

China’sChina’s carbon carbon emissions emissions reduction reduction policy policy started started from 2007, from when 2007, the state when council the statepromulgated council itspromulgated National Climate its National Change Climate Program, Change the first Program, national the response first national to climate response change to of climate any developing change of country.any developing The policy country. proposes The a diversified policy proposes energy development, a diversified andenergy renewable development, energy is and defined renewable as an importantenergy is part defined of the as national an important energy strategy. part of In the line national with the energy Paris Agreement, strategy. In China line claimed with the to have Paris reachedAgreement, its carbon China emissions claimed peak to have as early reached as possible its carbon and emissions is committed peak to aas 60–65% early as decreased possible energy and is consumptioncommitted to pera 60 unit–65% GDP decreased by 2030 energy relative consumption to 2005 (Source: per unit China GDP by Energy 2030 relative Outlook to 2030 2005 Report).(Source: TheChinaNational Energy Energy Outlook Development 2030 Report Strategy). The National Action Plan Energy (2014–2020) Development[38] issued Strategy by Action the State Plan Council (2014– states2020) that[38] China’sissued by non-fossil the State fuels, Council natural states gas, that and China’s coal willnon- accountfossil fuels, for 15%,natural 10%, gas, and and 62%, coal respectively, will account offor total 15%, primary 10%, and energy 62%, consumption.respectively, of The totalChina primary Energy energy Outlook consumption. 2030 Report The[39 China] even Energy claimed Outlook that domestic2030 Report primary [39] even energy claimed consumption that domestic would continue primary to energy be optimized consumption in response would to the continue pressure to of be reducedoptimized carbon in responseemissions toas well the as pressure environment of reduced constraints. carbon The emissions proportion as of coal well consumption as environment will decreaseconstraints. to 60% The with proportion a 17% share of coal of non-fossilconsumption energy willby decrease 2020. The to 60% proportion with a 17% of coal share consumption of non-fossil in 2030energy will by be 2020. further The reduced proportion to 49%, of coal while consumption non-fossil energy in 2030 is plannedwill be further to be 22% reduced at that to time. 49%, while non-Althoughfossil energy it is is important planned toto paybe 22% attention at that to time. the optimization of energy structure and international cooperation,Although the exploration it is important and application to pay attention of low-carbon to the economic optimization development of energy is still structure a challenging and task.international Continuing cooperation, the existing the energy exploration use pattern and application will not achieve of low the-carbon proposed economic emission development reduction is targets,still a challenging and the policies task. that Continuing have been the implemented existing energy already use mostly pattern advocate will not low-carbon achieve development.the proposed Theemission absence reduction of a concrete targets, policy and framework the policies needs that urgenthave been attention. implemented In consideration already ofmostly this, weadvocate focus low-carbon development. The absence of a concrete policy framework needs urgent attention. In consideration of this, we focus on four levels: operations, governance, institution, and culture (OGIC) to specify an alternative energy use paradigm shift and realize the emission reduction targets proposed internationally and domestically (see Figure 2).

Sustainability 2018, 10, 1639 5 of 18 on four levels: operations, governance, institution, and culture (OGIC) to specify an alternative energy use paradigm shift and realize the emission reduction targets proposed internationally and

Sustainabilitydomestically 2018 (see, 10, Figurex FOR PEER2). REVIEW 5 of 17

Figure 2. Operational, governance, institutional, and cultural (OGIC) analysis of the energy use Figure 2. Operational, governance, institutional, and cultural (OGIC) analysis of the energy use paradigm shift. paradigm shift. 3.1. Operations 3.1. Operations At the operational level, the energy use paradigm shift consists of policies orientated in relation At the operational level, the energy use paradigm shift consists of policies orientated in relation to the energy carbon emissions coefficient, energy intensity, and urbanization. to the energy carbon emissions coefficient, energy intensity, and urbanization. 3.1.1. Energy Carbon Emissions Coefficient 3.1.1. Energy Carbon Emissions Coefficient The energy carbon emissions coefficient (C/E) refers to the amount of carbon dioxide released The energy carbon emissions coefficient (C/E) refers to the amount of carbon dioxide released per per unit of primary energy consumption. China has been working decisively towards a market unit of primary energy consumption. China has been working decisively towards a market economy economy since 1992. In the ensuing year, many industrial parks sprang up, inducing a large since 1992. In the ensuing year, many industrial parks sprang up, inducing a large proportion of coal proportion of coal consumption to support the extensive growth. But now, we have entered a new consumption to support the extensive growth. But now, we have entered a new era that advocates a era that advocates a much lower coefficient to meet the requirements of sustainability. Enterprises much lower coefficient to meet the requirements of sustainability. Enterprises need to be encouraged need to be encouraged to tap into new energy markets and reduce their share of coal consumption. to tap into new energy markets and reduce their share of coal consumption. Although it may well be Although it may well be unrealistic to attempt to reduce the proportion of coal by a big margin in the unrealistic to attempt to reduce the proportion of coal by a big margin in the short term, there are still short term, there are still some opportunities for small-scale reductions in the proportions of both some opportunities for small-scale reductions in the proportions of both coal and oil. coal and oil. 3.1.2. Energy Intensity 3.1.2. Energy Intensity Endeavors to reduce energy intensity (E/GDP) need to pay attention to the ecological transformationEndeavors of to industrial reduce energy energy intensity consumption (E/GDP) and need the life to paycycle attention theory of to building the ecological energy transformationconsumption. Compared of industrial with theenergy advanced consumption techniques and used the in life developed cycle t countries,heory of building material energy consumption.production consumption Compared is with enormous. the advanced In China, techniques for example, used there in is developed 30–40% more countries, domestic building energy materialconsumption production than theconsumption general level is enormous. abroad. ToIn China, reduce for industrial example, energy there is consumption, 30–40% more domestic technical energyimprovements consumption and life-cycle than the energy general conservation level abroad. are To of great reduce importance, industrial where energy life consumption, cycle refers technicalto the whole improvements process of productionand life-cycle [40 energy,41]. Building conservation life cycle are of energy great consumptionimportance, where not only life takescycle refersconsumption to the whole during process building of production operation into [40, account,41]. Building but also life considerscycle energy the consumption preparation stage not only and takesconstruction consumption phase. during building operation into account, but also considers the preparation stage and construction phase. 3.1.3. Urbanization 3.1.3. Urbanization Life and production are two major supports for the paradigm shift. For life, the energy use paradigmLife and shift production requires energy are two saving major in supports daily life, for while the for paradigm production, shift. the For energy life, the use energy paradigm use paradigmshift means shift the requires evolution energy of industrial saving structure.in daily life, Urbanization while for production, has brought the more energy floating use populationparadigm shiftinto themeans cities, the many evolution of which of industrial work for structure. non-agricultural Urbanization industries, has brought especially more secondary floating population industries, intoand thusthe cities, promotes many the of manufacturingwhich work for boom. non-agricultural However, there industries, is a long-term especially cointegration secondary relationship industries, andbetween thus secondary promotes industries the manufacturing and energy consumption boom. However, [42,43]. Thethere most is effectivea long-term way to cointegration alleviate the relationshippressure of energy between demand secondary is by promotingindustries and industrial energy structure consumption evolution, [42,43 accelerating]. The most modern effective service way to alleviate the pressure of energy demand is by promoting industrial structure evolution, accelerating modern service industry development to reduce the over-reliance of the national economy on industrial growth and change the present mode of extensive growth.

3.2. Governance Governance in general is a set of management systems in relation to incentives and constraints. As for industrial production, the main interested parties are the government, the National People’s

Sustainability 2018, 10, 1639 6 of 18 industry development to reduce the over-reliance of the national economy on industrial growth and change the present mode of extensive growth.

3.2. Governance Governance in general is a set of management systems in relation to incentives and constraints. Sustainability 2018, 10, x FOR PEER REVIEW 6 of 17 As for industrial production, the main interested parties are the government, the National People’s Congress (NPC) (NPC) and and enterprises enterprises (or (or individuals) individuals) (see (see Figure Figure 3). 3Enterprises). Enterprises are the are center the center of micro of- economicmicro-economic action. action.A low-carbon A low-carbon economy economy cannot be cannot separated be separated from the frommarket the and market neither and can neither low- carboncan low-carbon enterprises. enterprises. The transformation The transformation process process involves involves externalities externalities of such of such public public goods goods as resourcesas resources and and the theenvironment, environment, which which results results in assumptions in assumptions concerning concerning the price, the price, output, output, cost, cost,and entryand entry barrie barriersrs cannot cannot meet meet the themarket market requirements requirements and and so soit is it isdifficult difficult for for the the market market to to effectively effectively allocate resources and hence resulting in eventual market failure. Therefore, Therefore, the paradigm shift must must be government-orientatedgovernment-orientated inin ChinaChina atat least. least. The The state state formulates formulates long-term long-term planning planning to to strengthen strengthen its itsguidance guidance for for enterprises. enterprises. Specific Specific policies policies could could be be creating creating energy-related energy-related standardsstandards and incentive measures to to encourage encourage enterprises enterprises to to join join the the low low-carbon-carbon development development mode. mode. In Inaddition, addition, NPC, NPC, a cruciala crucial participant participant in in the the system, system, should should play play a arole role in in supervision supervision by by the the government. government. The The energy energy use paradigm paradigm shift shift binding binding force force is weak is weak in China in China and the and local the government local government is insufficiently is insufficiently motivated motivatedto force the to issue. force A the monitoring issue. A monitoring mechanism mechanism is therefore is necessary therefore for necessary the integrity for the of theintegrity governance of the governancesystem of the system low-carbon of the energylow-carbon use paradigmenergy use shift. paradigm shift.

Figure 3. Incentive system for energy transformation. Figure 3. Incentive system for energy transformation.

3.3. Institution 3.3. Institution Why does coal occupy such a huge proportion in energy consumption in China? Firstly, coal is Why does coal occupy such a huge proportion in energy consumption in China? Firstly, coal is stored in large quantities. Moreover, it is part of related industrial chains, including petrochemical, stored in large quantities. Moreover, it is part of related industrial chains, including petrochemical, heavy industry, steel mills, and heating, which depend exclusively on coal. The most difficult and heavy industry, steel mills, and heating, which depend exclusively on coal. The most difficult and important aspect of coal consumption is “military restructuring”, i.e., reducing coal consumption important aspect of coal consumption is “military restructuring”, i.e., reducing coal consumption through the control of industries and regions. For the former, enterprises meet the industry’s coal through the control of industries and regions. For the former, enterprises meet the industry’s coal limit target by technological transition while, for the latter, the local government imposes a wealth limit target by technological transition while, for the latter, the local government imposes a wealth accumulation-led tax instead of the usual production-led as the main source of local revenue. accumulation-led tax instead of the usual production-led as the main source of local revenue. Industrial Industrial parks can provide considerable revenue for cities, which is why the local government parks can provide considerable revenue for cities, which is why the local government relentlessly relentlessly leases industrial land at a very low price. However, coal consumption rises as industrial leases industrial land at a very low price. However, coal consumption rises as industrial land spreads. land spreads. If the old tax mode could be disbanded, the development zone’s contribution to the If the old tax mode could be disbanded, the development zone’s contribution to the economy would economy would decline, which, in turn, would benefit the development of a low-carbon economy. decline, which, in turn, would benefit the development of a low-carbon economy. 3.4. Culture During periods of industrial civilization, the outburst of productivity triggered by science provided a historic jump for humankind and was a time when people felt they could conquer nature. The situation now is different, with the destruction of forests, environmental pollution, and all threating human survival. In aiming for sustainable development, it is important to recognize that we are an integral part of nature and the need for harmonization instead of conquest. The establishment of an environmental justice value system is particularly required for energy use paradigm shift. It is vital that future urban growth pursues social welfare other than economic progress. As with optimizing the urban and rural resources allocation system, it is important to

Sustainability 2018, 10, 1639 7 of 18

3.4. Culture During periods of industrial civilization, the outburst of productivity triggered by science provided a historic jump for humankind and was a time when people felt they could conquer nature. The situation now is different, with the destruction of forests, environmental pollution, and desertification all threating human survival. In aiming for sustainable development, it is important to recognize that we are an integral part of nature and the need for harmonization instead of conquest. The establishment of an environmental justice value system is particularly required for energy use paradigm shift. It is vital that future urban growth pursues social welfare other than economic progress. As with optimizing the urban and rural resources allocation system, it is important to coordinate social contradictions and ease urban-rural conflict [44,45] to steadily promote the energy use paradigm shift to low-carbon emissions.

4. Carbon Emissions Forecast: Three Simulations China’s carbon emissions play a crucial role in the world. In the context of the existing energy use pattern and energy use paradigm shift, we take China as a research focus to forecast its exact carbon emissions in 2030 based on the combination of Kaya identity and Monte Carlo simulation.

4.1. Reference Simulation The reference simulation denotes the forecast result through a simple linear fit with carbon emissions data from 2000 to 2015, with fitting function:

y = 532.63 ∗ (x − 2000) + 3443.2, R = 0.984.

When x = 2030, for example, China’s 2030 carbon emissions are predicted to be 19,422 MMtons. However, a single variable linear regression is not able to provide a precise predicted value, and is therefore only taken to be a reference value.

4.2. Comparative Simulations Two additional and different sets of comparative simulations are used in order to show the influence of the energy use paradigm shift more intuitively. Comparative simulation I focuses on the values of the coefficients in an environment with no paradigm shift, and simulation II considers the impact of the energy use paradigm shift on the parameter values. Specifically, energy policy has less influence on such parameters as per capita GDP, urbanization, and population. The values of these parameters are therefore the same in both simulations, while the values of the carbon emissions coefficient and energy consumption intensity are different.

4.2.1. Energy Carbon Emissions Coefficient (C/E) The alteration or amelioration of a local energy structure is reflected in the C/E variation. As mentioned above, the total energy consumption in China in 2015 was 4300 Mtce, of which coal, oil, natural gas, and other primary energy sources accounted for 64%, 18%, 6%, and 12%, respectively, with 10,541 MM tons of carbon emissions. The C/E in 2015 is therefore 2.46. Under the guidance of the energy use paradigm shift, the future Chinese energy structure can have a clear optimizing tendency, in which the contribution of coal consumption to the total amount of energy use continuously decreases, while natural gas and other primary energy sources increases. Combined with the reduction targets proposed in the official documents [38,39],the proportion of coal consumption in 2030 will be further reduced to 49%, while non-fossil energy is planned to be 22% at that time. Oil consumption is expected to have an estimated appropriate value of 17% and natural gas is 10%. Sustainability 2018, 10, 1639 8 of 18

Figure4 summarizes the C/E value of previous years and the predicted C/E value of the coming years. There are no obvious laws that describe the trend of previous C/E values. All that can be found is that the value is generally stable at 2.45 ± 0.1; thus we assign 2.45 to be the estimated value for the year after 2015 in comparative simulation I. In comparative simulation II, because of the implementation of the paradigm shift, the local government imposes a wealth accumulation-led tax instead of the usual production-led as the main source of local revenue and pursues social welfare other than economic progress, energy saving has become a binding indicator in industry; the future coefficient therefore has a gradual downward trend. Lin and Liu [18] estimate this as an average annual growth rate of −0.4% during 2016–2020. This being the case, the coefficient in 2020 would be 2.39. In view of the more stable energy structure, and growing difficulties in making reductions in the later period of development, we adopt −0.2% as the new growth rate from 2021 to 2030. This means thatSustainability the C/E 2018 in 2030, 10, x willFOR bePEER 2.34. REVIEW 8 of 17

2.60 Comparative I Comparative II 2.55

2.50

2.45

2.40

2.35

2.30 2000 2005 2010 2015 2020 2025 2030

FigureFigure 4. 4.Estimates Estimates of of the the energy energy carbon carbon emissions emissions coefficients coefficients for for 2000–2030. 2000–2030 (*. ( Predicted* Predicted energy energy carboncarbon emissions emissions coefficient coefficient (C/E) (C/E values) values after after 2015). 2015).

4.2.2.4.2.2. Energy Energy Intensity Intensity (E/GDP) (E/GDP) E/GDPE/GDP mainly mainly reflects reflects the the level level of of energy energy efficiency. efficiency. Excessive Excessive consumption consumption of of coal coal inhibits inhibits energyenergy efficiency efficiency improvement. improvement. In addition, In addition, the tertiary the tertiary industries industries can play cana greater play role a greater in promoting role in efficiencypromoting than efficiency the secondary than industries.the secondary Studies industries. have shown Studies that have industrial shown structure that industrial adjustment structure with theadjustment policy of backwith three the binarypolicy (suppressof back three the binary secondary (suppress industries the secondary and develop industries the tertiary and industries)develop the willtertiary improve industries) energy efficiencywill improve to some energy extent efficiency [46]. to some extent [46]. AsAs mentioned mentioned above, above, China China strived strived to to reach reach its its carbon carbon emissions emissions peak peak as as early early as as possible possible and and isis committed committed to to decreasing decreasing E/GDP E/GDP by by 60–65% 60–65% by by 2030 2030 relative relative to to 2005. 2005. Energy Energy consumption consumption in in 2005 2005 inin China China was was 2614 2614 Mtce Mtce with with a correspondinga corresponding GDP GDP of of CNY18.49 CNY18.49 trillion, trillion, and and thus thus E/GDP E/GDP should should be be 1.411.41 thousand thousand tons tons coal coal equivalent/CNY equivalent/CNY billion billion (Ttce/B). (Ttce/B). If If the the share share in in reduction reduction is is 60%, 60%, the the value value of of energyenergy consumption consumption in in 2030 2030 will will be be 0.564 0.564 Ttce/B, Ttce/B, or or 0.494 0.494 Ttce/B Ttce/B if if 65%. 65%. Accordingly, Accordingly, we we choose choose the the medianmedian of of 0.53 0.53 Ttce/B Ttce/B as as the the intensity intensity value value of of energy energy consumption consumption in in 2030. 2030. BasedBased on on the the data data of theof the total total energy energy consumption consumption and grossand gross national national product product from thefrom past the year, past theyear, E/GDP the E/GDP is calculated is calculated and the and fitted the fitted curve curve obtained. obtained. To guarantee To guarantee the accuracy the accuracy of the of forecasts, the forecasts, we disregardwe disregard the data the beforedata before 2009 for2009 they for show they show a significantly a significantly different different distribution distribution from those from after those 2009. after Combining2009. Combining the historic the historic data from data 2009–2015 from 200 and9–2015 estimator and estimator of 2030, of gives: 2030, gives: 푦 = 0.0042 ∗ (푥 − 2009)2 − 0.0806 ∗ (푥 − 2009) + 0.9552, 푅² = 0.997. y = 0.0042 ∗ (x − 2009)2 − 0.0806 ∗ (x − 2009) + 0.9552, R = 0.997. The linear fit shows it will reach the lowest value in 2019 (Figure 5). In the absence of the energy useThe paradigm linear fit shift, shows it will it will not reach further the decline lowest whenvalue inenergy 2019 consumption(Figure5). In intensity the absence reaches of the the energyminimum use paradigmvalue. Instead, shift, it it is will most not likely further to maintain decline whenits lowest energy level consumption or start to increase. intensity In reachesthis way, thethe minimum energy consumption value. Instead, intensity it is most of comparative likely to maintain simulation its lowest I is taken level as or 0.57 start for to the increase. coming In years. this For comparative simulation II, with the implementation of the energy use paradigm shift, energy- related standards and incentive measures will be created to encourage enterprises to join the low- carbon development mode and thus, the value of E/GDP is expected to decline further. We therefore assume this parameter will have a constant growth rate from 0.58 Ttce/B in 2020 to 0.53 Ttce/B in 2030.

Sustainability 2018, 10, 1639 9 of 18 way, the energy consumption intensity of comparative simulation I is taken as 0.57 for the coming years. For comparative simulation II, with the implementation of the energy use paradigm shift, energy-related standards and incentive measures will be created to encourage enterprises to join the low-carbon development mode and thus, the value of E/GDP is expected to decline further. We therefore assume this parameter will have a constant growth rate from 0.58 Ttce/B in 2020 to 0.53Sustainability Ttce/B 2018 in 2030., 10, x FOR PEER REVIEW 9 of 17

1.60 Comparative I Comparative II 1.40

1.20

1.00

0.80

0.60

0.40 2000 2005 2010 2015 2020 2025 2030

FigureFigure 5. 5.Estimates Estimates of of energy energy intensity intensity for for 2000–20302000–2030 (*(* PredictedPredicted energyenergy intensity (E/GDP) (gross (gross domestic product) values after 2015).domestic product) values after 2015).

4.2.3.4.2.3. UrbanUrban andand RuralRural PerPer CapitaCapita GDPGDP UrbanUrban perper capitacapita GDPGDP representsrepresents thethe grossgross domesticdomestic productproduct producedproduced byby perper unitunit ofof urbanurban population.population. There There are are no no official official figures figures of ofthe the exact exact number number of urban of urban or rural or ruralper capita per capitaGDP. Here, GDP. Here,we utilize we utilize the ratio the that ratio takes that takesaccount account of “urban of “urban per capita per capita disposable disposable income*urban income*urban population” population” and and“rural “rural per capita per capita disposable disposable income*rural income*rural population” population” to embody to embody the value the of value GDPu/GDPr. of GDPu/GDPr. Finally, Finally,we obtain we obtainthe urban/rural the urban/rural per capita per capitaGDP indirectly GDP indirectly by making by making use of use the of urbanization the urbanization rate rate and andpopulation. population. InIn thethe wakewake ofof urbanization,urbanization, thethe urbanurban andand ruralrural areasareas appearappear to be closer than the previous era. TheThe economiceconomic systemssystems inin bothboth areasareas havehave undergoneundergone profoundprofound changeschanges andand bothboth areare beginningbeginning toto experienceexperience some competition. competition. The The statistics statistics reveal reveal a substantial a substantial economic economic gap gap between between the theurban urban and andrural rural areas areas since since 2000. 2000. The Theaverage average urban urban per capita per capita disposable disposable income income in 2000 in 2000 was wasCNY CNY 6280 6280 and andCNY CNY 2253 2253 for urban for urban and andrural rural residents residents,, respectively. respectively. Fifteen Fifteen years years later, later,this had this changed had changed to CNY to CNY31,195 31,195 and CNY and 11,422 CNY, 11,422, respectively, respectively, a ratio of a ratio2.73:1. of Supposing 2.73:1. Supposing that y = (urban that y per = (urban capita disposable per capita disposableincome*urban income*urban population)/(rural population)/(rural per capita disposable per capita disposableincome*rural income*rural population), population), and then the and values then theof yvalues in 2000 of y and in 2000 2015 and are 2015 1.58 are and 1.58 3.49 and, 3.49,respectively. respectively. This This seems seems to to run run counter counter to to the the “new “new urbanizationurbanization policy”policy” proposedproposed inin China,China, whichwhich aimsaims toto promotepromote aa “people-oriented”“people-oriented” urbanization,urbanization, facilitatefacilitate the the fusion fusion development development of of rural rural primary, primary, second second and and tertiary tertiary industries industries and and finally finally narrow narrow the gapthe between gap between urban urbanand rural and development. rural development. Nevertheless, Nevertheless, it seems moreit seems likely more that the likely urbanization that the rateurbanization will reach rate 70%, will as reach increasing 70%, as numbers increasing of people numbers move of people to the move cities to createthe cities wealth. to create Assuming wealth. thatAssuming the efficiency that the of efficiency the urban of population the urban inpopulation creating GDP in creating is 50% GDP higher is than50% higher that of ruralthan that population, of rural thenpopulation, in 2030 wethen can in obtain:2030 we can obtain: 푦 = (3.49/1.58) ∗ 3.49 ∗ (1 + 50%) ≈ 11. y = (3.49/1.58) ∗ 3.49 ∗ (1 + 50%) ≈ 11. There are many ways of predicting GDP. Considering the variability of the economic environment,There are three many assumptions ways of predicting are adopted GDP. Considering in doing this the in this variability paper to of predict the economic GDP in environment, 2030 China. threeAssumption assumptions I are: On adopted the basis in of doing the GDP this indata this from paper 2000 to to predict 2014, GDPthe fitted in 2030 function China. is: 푦 = 39259 ∗ (푥 − 2000) + 28237, 푅² = 0.954. Setting x = 2030 gives the total predicted GDP in 2030 of CNY 120 trillion. Assumption II: The US Department of Agriculture projects the GDP of the US in 2030 to be 24.8 trillion dollars, followed by China with 22.2 trillion dollars, or CNY149 trillion (US Projections for the 2030 World Economy Ranking, http://watchingamerica.com/WA/2015/04/28/us-projections-for-the- 2030-world-economy-ranking/). Assumption III: President Xijinping states that China’s annual growth rate should be no less than 6.5% from 2016–2020 and then to experience a slight decline to around 6.0%. By this criterion, China’s GDP in 2030 will be CNY 162 trillion (China sets GDP target at about 6.5%, omitting aim to

Sustainability 2018, 10, 1639 10 of 18

Assumption I: On the basis of the GDP data from 2000 to 2014, the fitted function is:

y = 39259 ∗ (x − 2000) + 28237, R = 0.954.

Setting x = 2030 gives the total predicted GDP in 2030 of CNY 120 trillion. Assumption II: The US Department of Agriculture projects the GDP of the US in 2030 to be 24.8 trillion dollars, followed by China with 22.2 trillion dollars, or CNY149 trillion (US Projections for the 2030 World Economy Ranking, http://watchingamerica.com/WA/2015/04/28/us-projections- for-the-2030-world-economy-ranking/). Assumption III: President Xijinping states that China’s annual growth rate should be no less thanSustainability 6.5% from 2018, 2016–2020 10, x FOR PEER and REVIEW then to experience a slight decline to around 6.0%. By this criterion,10 of 17 China’s GDP in 2030 will be CNY 162 trillion (China sets GDP target at about 6.5%, omitting aim to go higher,go https://www.livemint.com/Politics/GVRX1Uaoo7qSNPvnEFRj6M/China-sets-GDP-target- higher, https://www.livemint.com/Politics/GVRX1Uaoo7qSNPvnEFRj6M/China-sets-GDP- at-about-65-omitting-aim-to-go-high.htmltarget-at-about-65-omitting-aim-to-go-high.html). ). TheThe results results of of three three GDP GDP predictionspredictions along with with three three different different urban urban per per capita capita GDP GDP and and rural rural perper capita capita GDP GDP are are shown shown in in TableTable1 .1. On On thethe basisbasis ofof thesethese comparisons, we we take take th thee conservative conservative approachapproach and and adopt adopt the the predictionspredictions of the US Department Department of of Agriculture Agriculture for for further further curve curve fitting. fitting. Inputting all the 2000 to 2015 data and the 2030 data produces two equations. One for urban per Inputting all the 2000 to 2015 data and the 2030 data produces two equations. One for urban per capita GDP: capita GDP: 푦 = 5.5652 ∗ (푥 − 2000)2 + 3953 ∗ (푥 − 2000) + 7734.6, 푅² = 0.991 y = 5.5652 ∗ (x − 2000)2 + 3953 ∗ (x − 2000) + 7734.6, R = 0.991 and the other for rural per capita GDP: and the other for푦 rural = −5 per.125 capita∗ (푥 − 2000 GDP:)3 + 192.45 ∗ (푥 − 2000)2 – 410.33 ∗ (푥 − 2000) + 5090.7, 푅² = 0.999. y = −5.125 ∗ (x − 2000)3 + 192.45 ∗ (x − 2000)2 − 410.33 ∗ (x − 2000) + 5090.7, R = 0.999. Figure 6 summarizes the annual urban/rural per capita GDP results. Figure6 summarizes the annual urban/rural per capita GDP results. Table 1. Urban/rural per capita GDP estimates based on different assumptions.

Table 1. Urban/ruralGDP per capitaUrban GDP-Rural estimates Urban based on per different Capita assumptions.Rural per Capita Assumptions (Trillion Yuan) Ratio GDP (Yuan) GDP (Yuan) AssumptionsI GDP (Trillion120 Yuan) Urban-Rural11:1 Ratio Urban per Capita104 GDP,761.9 (Yuan) Rural per22 Capita,222.2 GDP (Yuan) III 120149 11:111:1 104,761.9130,079.4 27 22,222.2,592.6 II 149 11:1 130,079.4 27,592.6 IIIIII 162162 11:111:1 141,428.6141,428.6 30 30,000.0,000.0

140,000 40,000

120,000 35,000 30,000 100,000 25,000 80,000 20,000 60,000 15,000 40,000 10,000

20,000 5,000

0 0 2000 2005 2010 2015 2020 2025 2030

urban per capita GDP rural per capita GDP

FigureFigure 6. 6.Estimates Estimates of of urban urban per per capita capita GDP GDP and and rural rural per per capita capita GDP GDP for 2000–2030for 2000–2030 (* Predicted (* Predicted urban perurban capita per GDP capita and GDP rural and per rural capita per GDPcapita after GDP 2015). after 2015).

4.2.4.4.2.4. Urbanization Urbanization TheThe urbanization urbanization rate rate is is a a measuremeasure ofof urban development conditions. conditions. Of Of course, course, a higher a higher rate rate is is notnot necessarily necessarily better. better. The The rates rates in even in even the most the most developed developed countries countries are between are between 70–85% 70– [8547%] because [47] because the state still needs primary industries to support the stable development of both the economy and society. The vast majority of studies agree that China will have reached a stable stage of urbanization by 2030 although the specific urbanization rates differ. For example, Nations [48] believes 70% of the population in China will be citizens by 2030. The China Development Research Foundation suggests that it will take China 20 years to deal with its semi-urbanization problem and raise the urbanization rate to 65% in 2030. Other studies also mainly concentrate at 65% with a fluctuation of no more than 5% [49–51]. Note that if a linear fit is applied to the urbanization rate dating from 2000 to 2015, the rate is 85.9%, which is absurd since the momentum of regional development is certain to reduce when the urbanization rate is over 70% [52]. So we disregard this regression result and assume the urbanization

Sustainability 2018, 10, 1639 11 of 18 the state still needs primary industries to support the stable development of both the economy and society. The vast majority of studies agree that China will have reached a stable stage of urbanization by 2030 although the specific urbanization rates differ. For example, Nations [48] believes 70% of the population in China will be citizens by 2030. The China Development Research Foundation suggests that it will take China 20 years to deal with its semi-urbanization problem and raise the urbanization rate to 65% in 2030. Other studies also mainly concentrate at 65% with a fluctuation of no more than 5% [49–51]. Note that if a linear fit is applied to the urbanization rate dating from 2000 to 2015, the rate is SustainabilitySustainability85.9%, which 2018 2018, is10, 10, absurd x, xFOR FOR PEER PEER since REVIEW REVIEW the momentum of regional development is certain to reduce when1111 of of the17 17 urbanization rate is over 70% [52]. So we disregard this regression result and assume the urbanization raterate has has a a constant constant growth growth rate rate that that finally finallyfinally reaches reaches 70% 70% in inin 2030. 2030.2030. Figure Figure 7 77 summarizes summarizes this thisthis annual annualannual urbanizationurbanization rate. rate.

FigureFigure 7. 7. Estimates EstimatesEstimates of of urbanization urbanization for for 200 200 2000–203000––20302030 ( *( (* *Predicted Predicted urbanization urbanization rate rate after after 2015 2015).2015).).

4.2.5.4.2.5. Population Population WorldWorld population population growth growth over over the the next next 30 30 years years is is expected expected to to mostly mostly take take place place in in Africa Africa and and India,India, which which which is is isprojected projected projected to to tosurpass surpass surpass China China China to to become tobecome become the the world’s the world’s world’s most most most populous populous populous country country country in in 2028, 2028, in and 2028, and China’sChina’sand China’s urban urban urban population population population will will willincrea increa increasesese by by a bya further further a further 310 310 310 million million million to to reach to reach reach a a peak apeak peak of of ofapproximately approximately approximately 1 1 1 billionbillion in in 2030 2030 before before gradually gradually declining declining [ 4 [48488].].] . TheThe linear linear fit fitfit result resultresult of ofof the thethe population populationpopulation based basedbased on onon the the 2000 2000 to to 2015 2015 data data isis is 1.481.48 1.48 billionbillion billion (R(R (R22=0.98)=2=0.98) 0.98) in in 2030.2030. However, However, China China has has now now fully fully implemented implemented its its “two “two children children policy”, policy”, which which allows allows one one couple couple toto have havehave two twotwo childrenchildren children fromfrom from 2015 2015 2015 instead instead instead of of theof the the previous previous previous one-child one one-child-child policy. policy. policy. Qu, etQu, Qu, al. et [et53 al. ]al. sampling [ 5[533] ]sampling sampling survey surveysurveyof the fertilityof of the the fertility fertility desire desireof desire women of of women women in China in in China suggests China suggests suggests that the that that population the the population population will reach will will areach reach maximum a a maximum maximum in 2027 ininbefore 2027 2027 enteringbefore before entering entering a period a ofa period period negative of of negative growth.negative Thegrowth. growth. regression The The regression equation,regression therefore,equation equation, can,therefore, therefore, only estimate can can only only the estimatepopulationestimate the the before population population the peak, before before namely the the 2016–2027. peak, peak, namely namely For the 201 201remaining66––20272027. . period, For For the the we remaining remaining conservatively period, period, set 0.2% we we conservativelyconservativelyas the constant set set negative 0.2% 0.2% as as growth the the constant constant rate.Figure negative negative8 summarizes growth growth rate. rate. the Figure Figure final results.8 8 summarizes summarizes the the final final results. results.

FigureFigure 8. 8. Estimates Estimates of of population population for for 200 200 2000–203000––20302030 ( *( (* *Predicted Predicted population population after after 2015 2015).2015).).

4.3.4.3. Monte Monte Carlo Carlo Simulation Simulation AA fundamental fundamental requirement requirement of of a a forecast forecast process process is is to to be be robust, robust, i.e. i.e., ,it it is is able able to to give give a a reasonably reasonably goodgood output output even even if if the the inputs inputs are are a a little little inaccurate. inaccurate. To To achieve achieve this this goal, goal, we we r eplacereplace the the single single valued valued parameterparameter with with a a probability probability distribution. distribution. Each Each parameter parameter is is defined defined as as being being normally normally distributed. distributed. TheThe estimates estimates of of the the variables variables proposed proposed in in above above section section are are adopted adopted as as the the mean mean of of their their distributionsdistributions while while the the standard standard deviatiodeviationsns are are acquired acquired from from a a comparison comparison between between data data at at an an adjacentadjacent time time (see (see Table Table 2 2a,a,b).b). AllAll the the parameters parameters are are therefore therefore substituted substituted into into the the model. model. The The number number of of simulation simulation iterations iterations isis set set at at 100,000. 100,000. The The frequency frequency for for the the year year 2030 2030 is is presented presented in in Figure Figure 9 9a,a,bb as as an an example. example. After After collectingcollecting the the statistics statistics from from all all the the research research years, years, including including the the means means and and standard standard deviations, deviations, the the finalfinal results results are are summarized summarized in in Figure Figure 10, 10, which which shows shows two two clear clear inflection inflection points points in in the the emissions emissions

Sustainability 2018, 10, 1639 12 of 18

4.3. Monte Carlo Simulation A fundamental requirement of a forecast process is to be robust, i.e., it is able to give a reasonably good output even if the inputs are a little inaccurate. To achieve this goal, we replace the single valued parameter with a probability distribution. Each parameter is defined as being normally distributed. The estimates of the variables proposed in above section are adopted as the mean of their distributions while the standard deviations are acquired from a comparison between data at an adjacent time (see Table2a,b). All the parameters are therefore substituted into the model. The number of simulation iterations is set at 100,000. The frequency for the year 2030 is presented in Figure9a,b as an example. After collecting the statistics from all the research years, including the means and standard deviations, the final results are summarized in Figure 10, which shows two clear inflection points in the emissions in 2015 and 2024. The comparative simulations I and II predict 20,141 and 18,069 MMtons of carbon emissions in 2030–1500 MMtons less than the reference simulation estimate.

Table 2. (a) Distribution of variables. (b) Distribution of variables.

(a) Comparative Simulation I Comparative Simulation II Energy Consumption Energy Carbon Energy Consumption Year Energy Carbon Intensity Emissions Coefficient Intensity Emissions Coefficient Mean Std. Mean Std. Mean Std. 2016 2.45 0.60 0.01 2.44 0.01 0.60 0.02 20172.45 0.58 0.01 2.43 0.01 0.58 0.01 20182.45 0.57 0.01 2.42 0.01 0.57 0.01 20192.45 0.57 0.01 2.40 0.01 0.57 0.01 20202.45 0.57 0.01 2.39 0.01 0.58 0.01 20212.45 0.57 0.01 2.38 0.01 0.58 0.01 20222.45 0.57 0.01 2.38 0.01 0.57 0.01 20232.45 0.57 0.01 2.37 0.01 0.57 0.01 20242.45 0.57 0.01 2.37 0.01 0.56 0.01 20252.45 0.57 0.01 2.36 0.01 0.56 0.01 20262.45 0.57 0.01 2.36 0.01 0.55 0.01 20272.45 0.57 0.01 2.35 0.01 0.55 0.01 20282.45 0.57 0.01 2.35 0.01 0.54 0.01 20292.45 0.57 0.01 2.34 0.01 0.54 0.01 20302.45 0.57 0.01 2.34 0.01 0.53 0.01 (b) Urban per Capita GDP Rural per Capita GDP Urbanization Population Year Mean Std. Mean Std. Mean Std. Mean Std. 2016 72,407 1448 26,801 536 57% 1% 1383 277 2017 76,544 1531 28,554 571 58% 1% 1390 278 2018 80,692 1614 30,170 603 59% 1% 1397 279 2019 84,851 1697 31,617 632 60% 1% 1404 281 2020 89,021 1780 32,864 657 61% 1% 1411 282 2021 93,202 1864 33,882 678 62% 1% 1418 284 2022 97,394 1948 34,638 693 63% 1% 1425 285 2023 101,598 2032 35,103 702 64% 1% 1432 286 2024 105,812 2116 35,246 705 65% 1% 1439 288 2025 110,038 2201 35,036 701 66% 1% 1446 289 2026 114,275 2285 34,441 689 67% 1% 1454 291 2027 118,523 2370 33,432 669 68% 1% 1461 292 2028 122,782 2456 31,978 640 69% 1% 1458 292 2029 127,052 2541 30,048 601 70% 1% 1455 291 2030 130,079 2602 27,593 552 70% 1% 1452 290 * Energy carbon emissions coefficient: MMtons/Mtce; * Urban/Rural per capita GDP: yuan. Population: million. SustainabilitySustainability 20182018, 10, 10, x, 1639FOR PEER REVIEW 13 of13 18 of 17

(a) Carbon emissions: MMtons

(b) Carbon emissions: MMtons

Figure 9. (a) Carbon emission simulation results of comparative simulations I in 2030 (* Probability Figure 9. (a) Carbon emission simulation results of comparative simulations I in 2030 (* Probability meansmeans the the chance chance of of appearance. appearance. Frequency Frequency means thethe numbernumber of of times times it it appears appears (a (a total total of of 100,000 100,000 times)times)).). (b ()b Carbon) Carbon emission emission simulation simulation results results of comparativecomparative simulations simulations II II in in 2030 2030 (* Probability(* Probability meansmeans the the chance chance of ofappearance. appearance. Frequency Frequency means means the thenumber number of times of times it appears it appears.. (A total (A totalof 100,000 of times)100,000). times)).

Sustainability 2018, 10, 1639 14 of 18

Sustainability 2018, 10, x FOR PEER REVIEW 14 of 17

25,000 Reference Simulation Comparative Simulation I Comparative Simulation II 20,000

15,000

10,000 Carbon Emissions: MMtons Emissions: Carbon

5,000

0 2000 2005 2010 2015 2020 2025 2030 2035 Year

22,000 Reference Simulation Comparative Simulation I

20,000 Comparative Simulation II

18,000

16,000

Carbon Emissions: MMtons Emissions: Carbon 14,000

12,000

10,000 2015 2020 2025 2030 2035 Year

FigureFigure 10. 10.Carbon Carbon emissions emissions simulation simulation results. results. (* (* The The figure figure below below is is an an enlarged enlarged view view of of the the circled circled areaarea in in the the figure figure above). above).

4.4.4.4. Results Results and and Discussion Discussion TheThe reference reference simulation simulation in in the the case case study study describes describes the the extrapolation extrapolation of of a a simple simple single single linear linear fittedfitted result resultof of carboncarbon emissionsemissions inin 2030 that may not not fully fully reflect reflect future future carbon carbon emissions, emissions, which which is isaffected affected by by multi multi-factor-factor changes changes such such as asthe the local local energy energy structure, structure, population population and and the theeconomy. economy. The Theeconomic economic and and social social development development situation situation in in the the next next 15 15 years years is is likely likely to be better thanthan thatthat ofof the the pastpast 15 15 years, years, so so the the reference reference estimate estimate maybe maybe the the lowest lowest possible possible value value under under the the present present extensive extensive developmentdevelopment mode. mode. ThisThis meansmeans thatthat ifif China China stillstill maintains maintains itsits previousprevious growthgrowth pattern pattern andand energyenergy consumptionconsumption structure, structure, the the carbon carbon footprint footprint in 2030 in 2030 will willcertainly certainly exceed exceed 20,000 20,000 MMtons, MMtons, nearly nearlytwice thattwice of that 2015 of (10,642 2015 (10,642MMtons). MMtons). There are significant differences between the carbon footprints predicted by the two comparative simulations. The gap begins in 2020, especially after 2024, and becomes more

Sustainability 2018, 10, 1639 15 of 18

There are significant differences between the carbon footprints predicted by the two comparative simulations. The gap begins in 2020, especially after 2024, and becomes more accentuated thereafter. If there is no energy use paradigm shift, the carbon emissions will reach 20,141 MMtons in 2030. However, with the energy use paradigm shift, the estimated value could reduce to 18,069 MMtons. In the comparative simulation II, 2015 and 2024 are two inflection points that separate the Chinese carbon footprint into three stages, namely an extensive consumption period (2000–2015), early energy transition period (2016–2023), and a late energy transition period (2024–2030). Economic development has consumed a large number of resources and energy in China in recent times, as manifested the 2000–2015 extensive mode, leading to a continuous increase in carbon emissions from 2000 [54,55]. Since 2015, industrial restructuring and intensive energy use began to progress steadily under an energy-saving background. A carbon trading market is planned to start in 2017. Energy use paradigm shift lags while the economy is still in a prosperity stage, and therefore carbon emissions maintain a relatively strong growth momentum until the end of the stage. As for the late energy use paradigm shift period after 2024, the energy transformation policy will be fully implemented and, at the same time, both the urbanization rate and population will reach a peak and remain stable [56]. In this situation, the growth rate of carbon emissions slows down. As can be seen from Figure 10, carbon emissions will reach a maximum around 2029 and then no longer increase significantly. In short, it will take some time for the policy to achieve its effect after introduction; therefore, ensuring an improved realization of low carbon development requires accelerating the implementation of the energy use paradigm shift.

5. Conclusions Urban growth in developing countries is often accompanied by an increase in carbon emissions. This paper proposes an alternative energy use paradigm shift. The Kaya identity and Monte Carlo simulation are the research methods. Three simulations are conducted for better understanding the effects of the energy use paradigm shift. The results show that the carbon footprint will decline drastically in 2030 with the guidance of the new energy paradigm. The peak carbon emissions value is forecasted to appear around 2029. The study has significant policy implications. The new energy use paradigm should provide a comprehensive improved approach for countries where booming economic indicators induced by the long-term blind pursuit of economic growth conceals such serious problems as wasted resources and increasing carbon emission. However, the paper contains some uncertainties and limitations. Although the effect of policy on the forecast process is considered, only a simple linear/curve is used to forecast the mean value of each variable, and further work is needed to better model the variables. However, while it is hard to quantify the effect of the energy use paradigm shift on carbon emissions precisely, this paper provides the basis for future policy novelties in China and other similarly placed countries.

Author Contributions: Y.W. and J.L. conducted the model building and wrote the paper; L.S. contributed the comparison and discussion; M.S. provided suggestions and made revisions to the manuscript. Acknowledgments: This research is co-funded by the National Natural Science Foundation (Project No: 71373231) and the Fundamental Research Funds for the Central Universities (No.2-2050205-17-182). The paper was also completed with support from the Center for International Earth Science Information Network (CIESIN), Columbia University, USA. Conflicts of Interest: The authors declare no conflict of interest.

References

1. Wu, Y.; Shen, J.; Zhang, X.; Skitmore, M.; Lu, W. The impact of urbanization on carbon emissions in developing countries: A Chinese study based on the U-Kaya method. J. Clean. Prod. 2016, 135, 589–603. [CrossRef] 2. Wei, Y.-M.; Liu, L.-C.; Fan, Y.; Wu, G. The impact of lifestyle on energy use and CO2 emission: An empirical analysis of China’s residents. Energy Policy 2007, 35, 247–257. [CrossRef] Sustainability 2018, 10, 1639 16 of 18

3. Ahn, C.; Lee, S.; Peña-Mora, F.; Abourizk, S. Toward environmentally sustainable construction processes: The US and Canada’s perspective on energy consumption and GHG/CAP emissions. Sustainability 2010, 2, 354–370. [CrossRef] 4. Solomon, S.; Plattner, G.-K.; Knutti, R.; Friedlingstein, P. Irreversible climate change due to carbon dioxide emissions. Proc. Natl. Acad. Sci. USA 2009, 106, 1704–1709. [CrossRef][PubMed] 5. Nordhaus, W. Critical assumptions in the Stern Review on climate change. Sci. 2007, 317, 1682. [CrossRef] [PubMed] 6. Qiao, X.; He, Z. Understanding The Paris Agreement: A Double-layered Analysis Framework of Industry Transition and Carbon Emissions. China Rev. Political Econ. 2016, 7, 118–143. 7. Li, J.; Wang, X. Energy and climate policy in China’s twelfth five-year plan: A paradigm shift. Energy Policy 2012, 41, 519–528. [CrossRef] 8. Shahbaz, M.; Lean, H.H.; Shabbir, M.S. Environmental Kuznets Curve hypothesis in Pakistan: Cointegration and Granger causality. Renew. Sustain. Energy Rev. 2012, 16, 2947–2953. [CrossRef] 9. Jalil, A.; Mahmud, S.F. Environment Kuznets curve for CO 2 emissions: A cointegration analysis for China. Energy Policy 2009, 37, 5167–5172. [CrossRef] 10. Tevie, J.; Grimsrud, K.M.; Berrens, R.P. Testing the environmental Kuznets curve hypothesis for biodiversity risk in the US: A spatial econometric approach. Sustainability 2011, 3, 2182–2199. [CrossRef] 11. Castiglione, C.; Infante, D.; Smirnova, J. Is there any evidence on the existence of an environmental taxation Kuznets curve? The case of European countries under their rule of law enforcement. Sustainability 2014, 6, 7242–7262. [CrossRef] 12. Richmond, A.K.; Kaufmann, R.K. Is there a turning point in the relationship between income and energy use and/or carbon emissions? Ecol. Econ. 2006, 56, 176–189. [CrossRef] 13. Ibrahim, M.H.; Law, S.H. Social capital and CO2 emission—Output relations: A panel analysis. Renew. Sustain. Energy Rev. 2014, 29, 528–534. [CrossRef] 14. Shahbaz, M.; Ozturk, I.; Afza, T.; Ali, A. Revisiting the environmental Kuznets curve in a global economy. Renew. Sustain. Energy Rev. 2013, 25, 494–502. [CrossRef] 15. Stern, D.I.; Common, M.S.; Barbier, E.B. Economic growth and environmental degradation: The environmental Kuznets curve and sustainable development. World Dev. 1996, 24, 1151–1160. [CrossRef] 16. Stern, D.I. The Rise and Fall of the Environmental Kuznets Curve. World Dev. 2004, 32, 1419–1439. [CrossRef] 17. Zhang, C.; Zhou, Q. Study on China’s social and economic development under the conditions of greenhouse gas emission reduction. Henan Sci. 2012, 30, 1549–1553. 18. Lin, B.; Liu, X. China’s carbon dioxide emissions under the urbanization process: Influence factors and abatement policies. Econ. Res. J. 2010, 8, 66–78. 19. Dhakal, S. Urban energy use and carbon emissions from cities in China and policy implications. Energy Policy 2009, 37, 4208–4219. [CrossRef] 20. Yuan, J.; Kang, J.; Yu, C.; Hu, Z. Energy conservation and emissions reduction in China—Progress and prospective. Renew. Sustain. Energy Rev. 2011, 15, 4334–4347. [CrossRef] 21. Cong, R. An optimization model for renewable energy generation and its application in China: A perspective of maximum utilization. Renew. Sustain. Energy Rev. 2013, 17, 94–103. [CrossRef] 22. Martínez-Zarzoso, I.; Maruotti, A. The impact of urbanization on CO2 emissions: Evidence from developing countries. Ecol. Econ. 2011, 70, 1344–1353. [CrossRef] 23. Dhakal, S. GHG emissions from urbanization and opportunities for urban carbon mitigation. Curr. Opin. Environ. Sustain. 2010, 2, 277–283. [CrossRef] 24. Kaya, Y. Impact of Carbon Dioxide Emission Control on GNP Growth: Interpretation of Proposed Scenarios; Intergovernmental Panel on Climate Change/Response Strategies Working Group, May 1990. 25. O’Mahony, T. Decomposition of Ireland’s carbon emissions from 1990 to 2010: An extended Kaya identity. Energy Policy 2013, 59, 573–581. [CrossRef] 26. Duro, J.A.; Padilla, E. International inequalities in per capita CO2 emissions: A decomposition methodology by Kaya factors. Energy Econ. 2006, 28, 170–187. [CrossRef] 27. Wang, C.; Wang, F.; Zhang, H.; Ye, Y.; Wu, Q.; Su, Y. Carbon emissions decomposition and environmental mitigation policy recommendations for sustainable development in Shandong province. Sustainability 2014, 6, 8164–8179. [CrossRef] Sustainability 2018, 10, 1639 17 of 18

28. Allen, R.C. The British Industrial Revolution in Global Perspective; Cambridge University Press: Cambridge, UK, 2009; Volume 1. 29. Glaeser, E.L.; Kahn, M.E. The greenness of cities: Carbon dioxide emissions and urban development. J. Urban Econ. 2010, 67, 404–418. [CrossRef] 30. McDonald, R.I.; Green, P.; Balk, D.; Fekete, B.M.; Revenga, C.; Todd, M.; Montgomery, M. Urban growth, climate change, and freshwater availability. Proc. Natl. Acad. Sci. USA 2011, 108, 6312–6317. [CrossRef] [PubMed] 31. Mahadevan, S. Monte Carlo Simulation; Mechanical Engineering: Basel, Switzerland; Marcel Dekker: Basel, Switzerland, 1997; pp. 123–146. 32. New, M.; Hulme, M. Representing uncertainty in climate change scenarios: A Monte-Carlo approach. Integr. Assess. 2000, 1, 203–213. [CrossRef] 33. Lo, S.-C.; Ma, H.-W.; Lo, S.-L. Quantifying and reducing uncertainty in life cycle assessment using the Bayesian Monte Carlo method. Sci. Total Environ. 2005, 340, 23–33. [CrossRef][PubMed] 34. Pozzi, S.A.; Padovani, E.; Marseguerra, M. MCNP-PoliMi: A Monte-Carlo code for correlation measurements. Nucl. Instrum. Methods Phys. Res. Sect. A 2003, 513, 550–558. [CrossRef] 35. Sham, P.; Curtis, D. Monte Carlo tests for associations between disease and alleles at highly polymorphic loci. Ann. Hum. Genet. 1995, 59, 97–105. [CrossRef][PubMed] 36. Zhang, K. China’s role, challenges and strategy for the low carbon world. China Popul. Resour. Environ. 2008, 18, 1–7. 37. Zou, X.; Chen, S.; Ning, M.; Liu, Y. An empirical research on the influence factor of carbon emission in Chinese provincial regions. Ecol. Econ. 2009, 3, 34–37. 38. Council, T.S. National Energy Development Strategy Action Plan (2014-2020); The State Council: Beijing, China, 2014. 39. Council, T.S. China Energy Outlook 2030 Report; The State Council: Beijing, China, 2014. 40. Klöpffer, W. Life cycle assessment. Environ. Sci. Pollut. Res. 1997, 4, 223–228. [CrossRef][PubMed] 41. Wang, M.Q.; Han, J.; Haq, Z.; Tyner, W.E.; Wu, M.; Elgowainy, A. Energy and greenhouse gas emission effects of corn and cellulosic ethanol with technology improvements and land use changes. Biomass Bioenergy 2011, 35, 1885–1896. [CrossRef] 42. Sari, R.; Ewing, B.T.; Soytas, U. The relationship between disaggregate energy consumption and industrial production in the United States: An ARDL approach. Energy Econ. 2008, 30, 2302–2313. [CrossRef] 43. Lee, C.-C. Energy consumption and GDP in developing countries: A cointegrated panel analysis. Energy Econ. 2005, 27, 415–427. [CrossRef] 44. Ann, T.; Wu, Y.; Shen, J.; Zhang, X.; Shen, L.; Shan, L. The key causes of urban-rural conflict in China. Habitat Int. 2015, 49, 65–73. 45. Shan, L.; Ann, T.; Wu, Y. Strategies for risk management in urban–rural conflict: Two case studies of land acquisition in urbanising China. Habitat Int. 2017, 59, 90–100. [CrossRef] 46. Zhang, W. Energy Efficiency, Industrial Structure and Regional Economic Development Gap in China. J. Shanxi Financ. Econ. Univ. 2010, 10, 28–34. 47. Bank, W. World Development Report 2010: Development and Climate Change; The World Bank: Washington, DC, USA, 2010. 48. Nations, U. Department of Economic and Social Affairs. Population Division. Ageing 2013; UN: New York, NY, USA, 2013. 49. Shen, L.; Cheng, S.; Gunson, A.J.; Wan, H. Urbanization, sustainability and the utilization of energy and mineral resources in China. Cities 2005, 22, 287–302. [CrossRef] 50. Li, J.; Dong, X.; Shangguan, J.; Hook, M. Forecasting the growth of China’s natural gas consumption. Energy 2011, 36, 1380–1385. [CrossRef] 51. Yu, B.; Lu, C. Change of cultivated land and its implications on in China. Chin. Geogr. Sci. 2006, 16, 299–305. [CrossRef] 52. Northam, R.M. Urban Geography; John Wiley & Sons: Rogers Hoboken, NJ, USA, 1979. 53. Qu, Z.; Zhang, X.; Jin, Y. Demographic consequences of an immediate transition to a universal two-child policy. Popul. Res. 2014, 38, 3–17. Sustainability 2018, 10, 1639 18 of 18

54. Zhang, Y.-J.; Da, Y.-B. The decomposition of energy-related carbon emission and its decoupling with economic growth in China. Renew. Sustain. Energy Rev. 2015, 41, 1255–1266. [CrossRef] 55. Chang, C.-C. A multivariate causality test of carbon dioxide emissions, energy consumption and economic growth in China. Appl. Energy 2010, 87, 3533–3537. [CrossRef] 56. Deng, W.; Hu, S. Non-linear combined prediction model of medium and long-term civil vehicle population of China. J. Transp. Syst. Eng. Inf. Technol. 2010, 10, 103–108. [CrossRef]

© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).