Revealing Dynamic Effects Of China's Regional Coal Consumption Control Policy On Energy E ciency And Pollution Emissions


 Up to now, far too little attention has been paid to comprehensively assessing the impact of the coal consumption constraint policy (CCCP) on energy intensity and pollution emissions based on China cities level. The present study explores the impact of CCCP on energy intensity and pollution emissions in a panel of 73 Chinese cities using an SEM approach from 2005 to 2019. We find that the CCCP has significantly reduced SO2 and CO2 emission intensity and energy intensity by 0.1283%, 0.0747%, and 0.2493% respectively, but the CCCP effect on CO2 emission intensity and energy intensity has a lag of 3 years. Thus CCCP can effectively reduce emission intensity through industrial restructuring and technology advanced is the only effective way to reduce energy intensity. The study provides some valuable suggestions to enable the control of coal consumption.


Introduction
As the largest producer and consumer of coal globally, the coal-based energy structure has supported China's rapid economic development (Guo et al., 2020;Li et al., 2019). However, since 2011, China has consumed more coal than the rest of the world combined, and it emits about a quarter of the world's greenhouse gases, the largest share of any country (O'Meara, 2020), which limits the development of China's ecological environment (Jiang et al., 2018). In the face of economic growth and air quality challenges, China takes coal control as a core target of sustainable economic development and has adopted a series of policies and measures in recent years. In 2011, China introduced the "Energy Conservation and Emission Reduction Work Plan during the 12th FYP", which proposes to put into practice the coal consumption constraint policy (CCCP) in the pilot city. In 2016, China launched the additional coal capacity reduction policy "Notice on Coal Consumption Reduction and Substitution", expanding the pilot city to optimize the energy structure further and strengthen the implementation of the target to control total coal consumption. As a series of coal consumption constraint policies had been carried out for nearly a decade, there is an urgent need to quantitatively assess the impact of the policy to determine whether it has increased energy e ciency and reduced emissions.
As a typical command-and-control policy, the CCCP has fueled debates about its impact. It is widely accepted that coal control strategies are essential for improving air quality (Yang & Teng, 2018). However, CCCP affects air pollution, CO 2 emissions, economy, and energy intensity is still debatable. Some scholars predicted that CCCP will inevitably lead to a reduction in social welfare (Zhang et al., 2021) and economic development (Xiao et al., 2020) but could contribute to controlling environmental pollution emissions, such as SO2 and NOx, PM2.5, CO2 (Xiao et al., 2020). Conversely, there are also some researchers who believe that CCCP, a non-market driven policy, was technically infeasible (Shi et al., 2018), it not only allows for a signi cant increase in economic costs (Shi et al., 2018) but also led to a rise in pollution emissions and energy intensity (Guo et al., 2020), due to technical and cost barriers. As for the effect of CCCP on energy intensity, Shi et al. (2018) found that coal-to-gas policies have an insigni cant negative effect on energy intensity in the Jing-Jin-Ji region. Guo et al. (2020) point out that CCCP fails to reduce the energy intensity based on 289 cities in China.
In terms of the archival path of CCCP impact, extensive research has shown because of the substitution effect from coal to natural gas and electricity (Shou et al., 2020), CCCP can reduce SO2, NOx, and CO 2 emissions, and optimize energy structure (Chen & Chen, 2019; Lin & Jia, 2020).
But Guo et al. (2021) argued that CCCP, as command-and-control environmental regulations, mainly reduced CO 2 emissions by improving technology levels. To the best of our knowledge, there is no study that investigates the speci c archival mechanism of CCCP effect at city levels. A few relevant studies focusing on cities only cover several large cities (Chen and Chen, 2017;Meng et al., 2017), due to the scarce of city-level data that cannot be generalized to most Chinese cities. Although Guo et al. (2020) conducted multi-city studies using data from the existing China City Statistical yearbook, they ignored the effects of CCCP on the second set of pilot projects introduced in 2016. Moreover, Guo et al. (2020) used the total energy consumption of 289 cities in China, but the data obtained were estimated based on provincial data which were not rigorous, and they did not provide a speci c estimation method.
In this study, we provide the evaluation of the impact of CCCP on CO2 emissions, air pollution, and energy intensity based on realistic data from policy consecutive prospects. Furthermore, we employ an SEM model to track the speci c realization path of CCCP affecting CO2 emissions, air pollution, and energy intensity. The main contributions of this paper are as follows. As an Economy-Energy-Environment (3E) database is constructed for analysis. As far as city-level database limited is concerned, there is little literature on total energy consumption at the city level. To ll the data gap, we constructed an Economy-Energy-Environment database including economic, environmental, total energy consumption indicators, and consumption indicators by energy species from 2005-2019 based on the original data from urban statistical yearbooks. Due to the lack of data for some cities, 73 cities with complete data were counted for empirical analysis in this paper.
As far as policy relevance and city limited is concerned, this study uses timing-varying DID to examine the synergistic effects of two consecutive typical CCCP policies on CO 2 emissions, air pollution, and energy intensity based on 73 cities across the country. In contrast, most of the existing studies focus only on a single CCCP policy or in limited big cities using a Scenario method, which is insu cient and unrealistic, Comprehensive analysis based on realistic data reveals a broader and accurate picture of the effects of CCCP. Unlike previous studies that mainly explored the environmental and economic impacts of CCCP separately, this paper estimates the effects of CCCP from the intensity perspective, including greenhouse gas intensity, pollution emission intensity, and energy intensity, which helps us to effectively embody the e ciency of the policy implementation. For the very rst time, a moderate model is introduced to more precisely identify the channels affecting greenhouse gas intensity, pollution emission intensity, and energy intensity. This analysis con rms the speci c mechanism driving China's coal consumption. This helps policymakers to further formulate coal policies to balance the reduction of coal consumption with other desirable environmental effects.

Literature Review
There are a large number of published studies describing the effects of CCCP policies. From an economic perspective, the consequences of CCCP are unfavorable. Shi et al. (2018) pointed out that full compliance with CCCP would lead to a signi cant gap between supply and demand, resulting in a substantial increase in coal prices and economic costs. The effects of different policy instruments of the CCCP are signi cantly different on the macroeconomic system, but regardless of which policy tool is used, the CCCP will inevitably lead to a reduction in social welfare (Zhang et al., 2021). In contrast, (Li and Yao, 2020) demonstrated that CCCP contributed to energy conservation and carbon emission reduction, but only with a slight loss of economic e ciency. Chen and Chen (2019) focused on "2+26 cities" and predicted the environmental effect of CCCP in the building sector. The results showed that the policy could cause an increase in natural gas and electricity consumption, which could help reduce SO 2 and NO x emissions, and it is helpful for some less developed areas to achieve the goal of low carbon faster. ln gso2 it = β 0 + β 1 Dit + θ 1 ln(pgdp) it + θ 2 industry it + θ 4 ln(pop) it + θ 5 ln(tec) it + θ 6 ln(fdi) it + θ 7 coal __ c + μ i + γ t + ϵ it 1 ln gco2 it = β' 0 + β' 1 Dit' + θ' 1 ln(pgdp) it + θ' 2 industry it + θ' 4 ln(pop) it + θ' 5 ln(tec) it + θ' 6 ln(fdi) it + θ' 7 coal __ c + μ' i + γ' t + ϵ' Where i and t denote cities and years, respectively; gso2 is a measure of so2 emission intensity; gco2 represents co2 emission intensity; and toconsum_c denotes energy intensity that depends on pgdp (GDP per capita), industry (secondary industry output value), FDI (foreign direct investment), coal_c (coal consumption in total energy consumption), pop (population), and tec (technological progress). The variable of interest is D it , D it ′, D it ′′, dummies that take the value of 1 in the years after city i implementing CCCP and 0 otherwise. The coe cients β 1 , β' 1 , β' 1 , are the parameters to be estimated, representing the net effect of CCCP. A positive and signi cant one indicates the CCCP exerts a positive effect on the intensity of SO 2 emission and CO 2 emission, energy intensity, while a negative and signi cant one suggests that CCCP pushed intensity of SO 2 emission and CO 2 emission lower. While, μ i , μ′ i , μ′′ i and γ t , γ′ t , γ′ ′ t are vectors of city and year dummy variables that account for city and year xed effects, but ϵ it , ϵ′ it , ϵ′′ it are the random disturbance terms.
To further understand the speci c implementation path of the policy, we employ the intermediate model to test which path-among adjusting the industrial structure, promoting technological progress, and reducing energy intensity-is most effective in reducing CO 2 and SO 2 emissions and energy intensity. The traditional multiple mediation model, a regression analysis based on the causality test proposed by Baron and Kenny (1986), is less e cient in estimating mediation effects (Fritz and MacKinnon, 2007). The Sobel test used in regression analysis assumes that a*b obeys a normal distribution, whereas mediation effects often do not meet the required, resulting in test results being relatively unreliable (Bollen and Stine, 1990; Stone and Sobel, 1990). SEM can address the shortcomings of the above methods. It is the best framework for mediated effects analysis because it can estimate all model parameters simultaneously and e ciently (Zhao et al., 2010). Therefore, this study explores the speci c pathways through which CCCP works by building SEM models.

Data
This study examines the impact of CCCP on CO2 emissions, air pollution, and energy intensity by covering 73 cities of China over the period from 2005 to 2019. Table A1 reported the list of cities in the appendix. Therefore, CO 2 emission, energy structure, and energy intensity coal consumption data were obtained based on our unique database. Speci cally, we collected all data on 24 energy types published in the urban statistical yearbooks. Then, cities with missing for consecutive years were removed and the number of cities was nally set to 73. The raw energy data are uniformly converted to standard coal and summed up so that we get the total energy consumption. Additionally, as for coal consumption, it is calculated using 12 coal energies of the industry which is uniformly converted to standard coal and summed up. The rest of the data come from the China city statistical yearbooks. The variables description and descriptive statistics of the variables are shown in Table 1.

Results And Discussion
In Table 2, ndings of time-varying DID show that CCCP can signi cantly reduce the intensity of CO2 and SO2 emissions as well as energy intensity.
The nding indicates that CCCP has decreased so2 intensity by 0.1283% and co2 emission intensity 0.0747%. Such results demonstrate that the CCCP, developed with air pollution control as the starting point, can achieve synergistic control of greenhouse gases and air pollutants. There are two possible explanations are that fossil fuel combustion is a signi cant cause of air pollution and an important anthropogenic source of greenhouse gases. Conventional air pollutants such as SO2 have a high degree of structural similarity to CO2 emissions. Therefore, limiting coal consumption, which is the primary energy source in China, can simultaneously reduce CO2 and air pollutants emissions, leading to co-control effects for the climate and the environment. Findings infer that two different types of environmental problems, air pollution, and greenhouse gases, are mostly driven by the same energy production and consumption patterns and therefore have the same emission reduction pathways. The effect of the coal reduction policy on energy intensity is negative signi cant and indicates that the implementation of the CCCP signi cantly reduces energy intensity by -0.249%. This means that technological progress and economic development are the most critical factors in reducing energy intensity. Note: t statistics in parentheses. * p < 0.1, ** p < 0.05, *** p < 0.01. city and year represent time xed effects and individual xed effects Table 3 reported the results of robustness. We assume that the CCCP was implemented 2 or 3 years ahead of schedule. Speci cally, it is supposed that the rst CCCP proposal was implemented in 2008 and 2009 and that the pilot cities were expanded in 2013 and 2014, respectively. Therefore, variables D_2 and D_3 are constructed as virtual policy dummies to reexamine counterfactual event effects. From Table 3, it can be seen that the coe cients of the effects of virtual policy (D_2/D_3) are not signi cant for both virtual policy times, and this result supports the hypothesis of the randomness of the timing of policy implementation. Therefore, the previously obtained results are robust and reliable. Note: t statistics in parentheses. * p < 0.1, ** p < 0.05, *** p < 0.01.
The results of time-varying DID show that the CCCP can signi cantly reduce the carbon emission intensity, So2 emission intensity, and energy intensity of the pilot cities. To further investigate the speci c realization path of CCCP implementation to reduce CO2 and SO2 emission intensity and energy intensity, throw light on the effective initiatives of China. The SEM model is employed in this paper; we select technological progress, energy intensity, and industrial structure as mediating variables for mediation analysis according to the analysis of the mechanism of CCCP.
Furthermore, due to the nonlinear distributional properties of the mediation effects, a nonparametric bootstrapping method is used in this paper to adjust the estimation bias (MacKinnon et al. 2004).
The nding of Table 4 shows intermediary effects of co2 emission intensity. The con dence interval (BC and p interval 95%) of the direct effect contains 0, indicating that the direct effect of coal reduction policy on CO2 intensity is not signi cant and is a fully mediated effect. In addition, the coe cient of the mediated effect of CCCP → industrial structure → CO2 intensity is -2.629%, the con dence interval (BC interval 95%) does not contain 0, denoting that industrial restructuring is an important mechanism for CCCP to reduce co2 intensity. In contrast, the con dence interval of the mediating effect of coal reduction policy → technological progress → CO2 intensity and coal reduction policy → energy structure → CO2 intensity both contain 0, representing that CCCP does not effectively reduce CO2 intensity by technological progress and energy structure optimization. If the 95% con dence interval excludes zero, it means that the mediator is statistically signi cant at the 5% level, otherwise, it is not signi cant. emission intensity by industry structure adjustment. whereas the con dence interval of the mediating effect of CCCP → technological progress → so2 emission intensity and CCCP → energy structure → so2 emission intensity both contain 0, demonstrating that CCCP does not effectively reduce so2 emission intensity by industrial structure adjustment and energy structure optimization. Similarly, the estimation results of the indirect effects of energy intensity are presented in Table 6. The con dence interval for the direct effect (BC interval 95% and P interval 95%) contains 0, indicating that the direct suppression effect of coal reduction policy on energy intensity is not signi cant and it is a fully mediated effect. Additionally, the coe cient of the mediated effect of CCCP → technological progress → energy intensity is -5.389825%, and the con dence interval (BC interval 95%/ P interval 95%) does not contain 0. The CCCP was able to effectively reduce energy intensity through increased technological progress in industrial production, caused by the innovation compensation effect promoted by the command-and-control policy of the government. While the con dence interval of the mediating effect of coal production reduction policy → industry structure → energy intensity and coal production reduction policy → energy structure → energy intensity both contains 0, indicating that coal production reduction policy is not effective in reducing energy intensity through the industry structure adjustment and energy structure optimization. Our ndings show that CCCP can effectively reduce emissions by restructuring the industry, rather than in other ways. To facilitate the implementation of the CCCP policy, China has set a strict ratio of backward production capacity elimination and forced the closure of more than 1,000 backward coal mines. As a command-and-control policy, by setting a strict backward industry goal for local governments and enterprises, the effect of CCCP is ensured effectively. What is surprising is that China invests tons of nancial support to the enterprise to replace coal with other clean energy, but adjust energy structure fail to reduce emissions signi cantly. This may be due to the fact that the CCCP, as a command-andcontrol policy, encourages the reduction of coal quantity and the use of alternative energy sources, which likely leads to a loss of resource allocation In contrast, energy intensity, to which China attaches greater importance, can only be reduced through technological advances. Mandatory measures, such as restructuring the industrial and energy structure, have not played a signi cant role in reducing energy intensity.

Conclusions And Policy Implementations
This study examines the impact of energy e ciency and pollution emissions from the perspective of continuity of policies implemented in 2011 and 2016 based on the SEM method, using special panel data of 73 Chinese cities from 2005 to 2019. Our SEM ndings identi ed that CCCP has co-bene ts of reducing emissions intensity and optimizing energy e ciency. As expected, it has led to a reduction in China's So2 emission intensity, CO2 emission intensity, and energy intensity by 0.1283%, 0.0747%, 0.2493% respectively in the pilot cities compared to other cities. The study also shows that CCCP, as a command-and-control policy, is effective to curb emissions intensity by industrial restructuring, rather than in progressing technology and optimizing the energy structure. Whereas, technology advanced is the only effective way to reduce energy intensity.
The above ndings have signi cant policy implications. Firstly, greater efforts are needed to ensure the expected effectiveness and reduce the time required for policy action. The government should play a key role in coordinating the various issues involved and take steps to avoid policy fragmentation and reduce the cost of friction between the various institutions involved. Secondly, the optimization effect of industrial structure as a crucial mechanism to reduce emission intensity plays an important role in environmentally balanced development. Therefore, in adjusting the industrial structure with the support of government nancing for environmental protection, attention should be paid to raising the entry threshold for enterprises with high energy consumption and promoting and developing green industries with high added value. Finally, the synergistic mechanism of environmental governance and structural optimization should be used to revitalize the city's innovation momentum, increase support for emission reduction R&D technologies, and promote a green orientation of R&D investment.
There are some limitations to our research: This study uses data from the industrial sector only by ignoring other sectors. The limited data may lead to an underestimation of the effect of CCCP on environmental pollution. Although we are shifting our eye contact from a few major cities in the