Assessing the Implications of Ecological Civilization Pilots in Urban Green Energy Industry on Carbon Emission Mitigation: Evidence from China

Abstract

This study aims to explore the role of China’s Ecological Civilization Pilot Policies in carbon emissions reduction within the urban green energy industry. It further investigates how these policies influence carbon emissions. To achieve this, a unique incentive–constraint model is established considering China’s distinctive political system. The DID model was used in this study, employing Chinese city data spanning from 2009 to 2020 and analyzing urban panel data with the use of two specific policies as quasi-natural experiments. The analysis reveals the following key findings: (i) Ecological Civilization Pilot Policies in the energy industry substantially contribute to carbon emission reduction through the effects of technological progress and industrial structure optimization; (ii) the unique incentive–restraint mechanism within these policies enhances their effectiveness, with short-term incentives and carefully designed assessment criteria playing a pivotal role in their successful implementation. These findings carry substantial implications for shaping environmental policies within the energy industry, emphasizing the importance of such policies in the ongoing global effort to reduce carbon emissions and promote sustainability.

1. Introduction

Pilot policies represent a distinctive policy, particularly when applied to the experimental implementation of “ecological civilization”, a rarity on the global stage. The notion of ecological civilization constitutes a unique buzzword within China, having garnered substantial influence within the nation since its integration into the ideology of the Communist Party of China in 2007 [1]. While this concept’s definition bears a resemblance to that of “Ecological democracy” [2], China’s developmental trajectory has transformed it into a societal vision that places paramount importance on enhancing the well-being of the populace and advancing national development with a steadfast commitment to the principles of sustainable development [3]. Since 2013, the Chinese government has implemented Ecological Civilization Pilot (ECP) policies in energy industry fields which encompass various strategies such as upgrading the energy structure, developing green industries, and protecting the ecological environment. These policies have provided China with practical foundations and accumulated experience in reducing carbon emissions and taking climate action [4].

The primary objective of this study is to delve into the multifaceted impact of China’s ECP policies, implemented over a decade, on the reduction of regional carbon emissions. The interplay between these policies and regional carbon emission patterns is scrutinized, taking into consideration the unique incentive–restraint mechanisms that influence local officials. Empirical insights into the effectiveness of the ECP policy paradigm and how incentive and constraint mechanisms impact policy outcomes are sought after in this research.

Our research extends beyond the validation of existing theories, seeking to provide practical insights that have broader implications for global governance. The ongoing debate between neoclassical economics and the Porter Hypothesis, particularly within the context of environmental regulatory policies and their impact on carbon emissions, is navigated [5,6,7]. The distinctions in policy intensity, objectives, mechanisms, and assessments among various environmental regulatory policies are focused on, shedding light on the nuances of government policies and their varying effects. Furthermore, the potential to inform future policy formulation and promotion exists in our findings. The relationship between ECP policies, incentive–constraint mechanisms, and policy effects is examined, contributing to a theoretical foundation that can guide policymakers and researchers in their endeavors to address carbon emissions and ecological civilization.

The remainder of this paper proceeds as follows: Section 2 provides an overview of the theoretical underpinnings and establishes a hypothesis that will be utilized in the empirical examination. Section 3 describes the data and methodology. Section 4 provides the core results on the impact of ECP policies and discusses how incentive–constraint mechanisms influence these policies. Section 5 concludes this paper.

2. Theoretical Framework

2.1. ECP Policies Overview

In December 2013, China’s National Development and Reform Commission (NDRC) proposed the establishment of an “Ecological Civilization Demonstration Area” in 100 regions across China. In 2017, China initiated the development of “China’s Demonstration Cities & Counties for Ecological Civilization Construction”. Both pilot policies were implemented at the prefecture-level cities and exhibited significant differences in policy indicators. This divergence allows for an impartial analysis of the correlation between ECP policies and carbon emissions. For the purpose of experimentation, these policies will be referred to as Policy A1 and Policy A2, respectively. The specific details of these two policies can be found in

Table 1. ECP Policies.

Policy Details	Policy A1	Policy A2
Policy Name	China’s Demonstration Cities & Counties for Ecological Civilization Construction.	Ecological Civilization Demonstration Area
Policy Objective	Establish a model for ecological civilization construction.	Explore ecological civilization construction in pilot areas.
Entry Method	Application Review Process.	Application Review Process
Responsible Agency	Ministry of Environmental Protection.	National Development and Reform Commission, Ministry of Finance, Ministry of Land and Resources, Ministry of Water Resources, Ministry of Agriculture, State Forestry Administration.
Policy Characteristics	National-level green honor.	Selective experimentation.
Policy Indicators	Clear construction targets and management procedures.	Pilot areas determine policy indicators based on their local conditions.
Assessment Method	Assessment is conducted first and the title is awarded after passing the assessment.	Regular inspections with qualification cancellation for areas that fail acceptance after the five-year construction period.
Policy Coverage	Titles are awarded annually, covering a total of 262 cities and counties in 2020.	In 2014 and 2015, a total of 100 representative areas were selected.

.

Analyzing the content of the pilot policies reveals commonalities between Policy A1 and Policy A2. Both policies feature an application–review mechanism, whereby local governments are required to initiate the application process, culminating in central authorities determining the approved list following a comprehensive review. Furthermore, both policies entail the establishment of specific construction and assessment criteria.

Notably, these policies diverge in terms of their respective incentive assessment mechanisms, manifesting as follows:

• Incentive effects: Policy A1 exhibits a stronger actual incentive impact. As a form of regional “green honor”, Policy A1 represents a short-term honor. Local governments need to meet the nationally prescribed assessment criteria within a short time frame. Afterward, the Ministry of Ecology and Environment issues corresponding titles to regions that pass the assessment. These titles can be considered a local official’s green achievements. Therefore, local leaders strive to meet the assessment standards. In contrast, Policy A2, which involves long-term pilot programs, is influenced by the actions of local leaders. Currently, implementing this policy does not necessarily guarantee incentives for local officials. Furthermore, the central government has not specified how honors can be obtained under Policy A2, potentially leading local officials to lack motivation for such pilot programs.
• Constraint systems: Policy A1, serving as an “honorary title” for a city, has stricter evaluations due to clear assessment criteria. This process involves a one-time assessment; once the criteria are met, the title is conferred without subsequent evaluation phases. As a result, this policy may cause local governments to prioritize their economic interests and discontinue the specific ecological civilization development measures outlined during the application and approval process. On the other hand, Policy A2 is an exploratory pilot, where the assessment criteria are unclear. Yet, there is a follow-up evaluation mechanism, and local governments must face the possibility of forced withdrawal if they do not pass the assessment.

When considering the two policies comprehensively, it becomes apparent that Policy A1 exhibits higher policy intensity. This is because local governments are obligated to fulfill tasks assigned by the central government in accordance with established regulations before undergoing central government inspections. This aligns with the principles of the administrative subcontracting system and the theories related to promotion tournaments. In contrast, Policy A2 involves a broader range of departments in initiating pilot programs and allows for more discretion. Local governments under Policy A2 may not have a greater sense of urgency to complete these tasks compared to those under Policy A1. There are some areas where both policies are concurrently implemented. In these cities, local governments must ensure the effective implementation of Policy A1 to earn the honorary title for their cities while also meeting the objectives of Policy A2 to address certain issues. Consequently, it can be inferred that among all cities, those implementing both policies exhibit higher policy intensity compared to those implementing either policy individually.

2.2. Impact of ECP Policies on Carbon Emissions

We believe that the implementation of ECP policies has led to higher requirements for energy enterprises. To align with the objectives delineated by the central government, local administrations operating within the framework of the target responsibility system will necessitate a transformation in their governance approach. This transition entails the adoption of more stringent environmental regulations within their respective jurisdictions. Typically, these measures fall into two primary categories: incentives and punishments [8,9]. Incentives encourage enterprises to reduce carbon emissions through diverse methods, encompassing technological advancements and production reduction, thereby catalyzing a city-wide reduction in carbon emissions. Conversely, punishments intensify the consequences of polluting emissions, discouraging polluting companies from continuing environmentally harmful production practices solely for economic gain. This ultimately leads to a reduction in carbon emissions. In this scenario, the implementation of ECP policies requires energy enterprises to augment their investments in research and development (R&D) and innovation. This augmentation aims to cultivate environmentally sustainable and more efficient technologies. As a result, it reduces energy and resource waste, facilitating proficient production management and cost-saving measures. Consequently, this initiative culminates in the reduction of carbon emission intensity, an enhancement of carbon efficiency, and the promotion of sustainable development.

Hypothesis 1 (H1).

The implementation of ECP policies incentivizes urban areas to reduce carbon intensity and enhance carbon efficiency in the energy industry.

2.2.1. Energy Industrial Optimization

ECP policies typically align with the long-term carbon reduction objectives of the nation or region. When formulating these policies, the central government establishes the level of carbon emissions within the region. Local governments, upon receiving these tasks, continuously employ administrative measures to compel enterprises to align their actions with the long-term policy objectives. These environmental regulatory measures compel both low-end and high-energy-consuming enterprises to undertake industrial upgrades, thereby reducing pollution emissions during the production process [10,11]. Consequently, they optimize the industrial structure, expedite regional industrial transformation and enhancement, reduce carbon intensity, and enhance carbon efficiency.

The requirements for ecological environment development will lead the government to increase the costs of environmental governance within the government budget while implementing environmental regulatory policies. This will also result in corresponding expenditures on green public services [12,13], thereby promoting the improvement of the ecological environment in the region. The “pollution haven” theory suggests that differences in environmental regulations among various regions can impact the decision-making process of pollution-intensive industries [14]. Regions with strict environmental regulations may incur higher costs for energy enterprises due to environmental issues, which can result in increased production costs and reduced comparative advantages for their products. Conversely, regions with more relaxed environmental regulations may attract polluting enterprises by offering lower environmental costs, thus becoming “pollution havens”. Enterprises in the energy sector are categorized as being subject to stringent environmental regulations, consequently incurring significant costs associated with ecological and environmental management, thereby resulting in elevated production expenses. As the implementation of ecological civilization initiatives gains traction in a specific region, the corresponding policy imperatives stimulate improvements in the local ecological environment. This, in turn, creates a favorable external environment for the development of green technology-oriented industries, effectively mitigating the costs associated with industrial transformation. Concurrently, these policy-driven changes compel polluting enterprises to either exit the region or undergo significant industrial upgrades. This dynamic fosters a competitive environment in which regions strive to surpass each other in terms of environmental performance, resembling a “race to the top”. Ultimately, this leads to a significant increase in the regional industrial agglomeration effect, thereby reducing carbon intensity and improving carbon efficiency.

Hypothesis 2a (H2a).

The implementation of ECP policies incentivizes urban areas to reduce carbon intensity and enhance carbon efficiency by optimizing energy industrial structure.

2.2.2. Green Technological Progress

In the domain of technological progress, the implementation of ECP policies has a significant impact on the extent of environmentally friendly innovation in the area, primarily driven by two contrasting effects: the “compliance cost” and the “innovation compensation”.

The “compliance cost” signifies how ECP policies prompt local governments to recalibrate their expectations concerning the ecological environment. As these policies are enacted, local governments enhance relevant ecological standards and emphasize corporate behavior regarding energy conservation and emission reduction. This increases the “compliance cost” for businesses, resulting in a “crowding out effect” that displaces innovation inputs, leading to the outflow of capital and hindering technological advancement [15,16]. While the “compliance cost” compels enterprises to either transition or upgrade their industries, it simultaneously hinders the improvement of carbon efficiency within the region.

The concept of “innovation compensation” refers to the practice in which local governments provide financial incentives to businesses that achieve green innovation in line with the predefined objectives of the central government. Following the “Porter Hypothesis”, suitable environmental regulations induce businesses to internalize costs, thus propelling them to actively devise green processes, products, and technologies. By doing so, companies can not only alleviate the financial burden of policy implementation but also potentially generate additional revenue [17]. Initiated from the perspective of public choice theory at the enterprise level, this “innovation compensation” for green technological progress can rouse the subjective initiative of enterprises functioning as “rational economic agents”. This stimulation encourages them to invest in green technology innovation, leading to improved efficiency in the use of raw materials and energy, reduced operational costs, and the attainment of policy incentives [18]. Ultimately, this contributes to the achievement of environmental regulatory objectives, such as the reduction of carbon emissions and the enhancement of carbon efficiency.

Hypothesis 2b (H2b).

The implementation of ECP policies incentivizes urban areas to reduce carbon intensity and enhance carbon efficiency by promoting technological progress.

2.3. Incentive–Restraint Mechanism in ECP Policy

While countries worldwide have established environmental laws and policies [19,20,21], China’s ECP policy stands out for its distinctive political context. China’s current implementation of ECP policies predominantly adheres to a “local application, central oversight” model. Drawing from the theoretical perspective of “administrative subcontract” as expounded by Zhou [22], China’s governance model can be synthesized as a composite of “vertical subcontracting” and “horizontal competition”. The central government gradually delegates public administrative responsibilities and administrative discretion to subnational levels, empowering local administrative leaders with substantial authority. Concurrently, the power to appoint lower-level government leaders is vested in their higher-level “contractors”. This model encapsulates the gradual devolution of public administrative responsibilities and administrative discretion to local tiers, endowing local administrative leaders with significant influence. However, in the process of policy issuance at different levels, local governments, acting as both “agents” responsible for policy implementation and “self-interested” entities, may not always align their goals with those of the central government. This duality stems from local governments serving as agents for superior governments’ directives while concurrently pursuing their own political and financial interests.

To address this principal–agent problem between central and local governments, the central government must establish incentive and restraint mechanisms. We believe that this unique Chinese incentive–restraint mechanism impacts two opposing effects: “compliance cost” and “innovation compensation”. Evaluating how incentives and constraints affect carbon emissions and carbon efficiency under different policies is imperative for comprehending China’s pilot policies.

2.3.1. Incentive Mechanism

Local government officials’ career trajectories are often intricately linked to opportunities for advancement. In the Chinese political system, there is a strong emphasis on performance-oriented promotions, with outstanding achievements being a prerequisite for ascending to higher-level government positions. As a result, local officials are often required to demonstrate their worth through notable accomplishments in order to stand out in the competition for government positions. In recent years, China has increasingly prioritized the development of a “green economy” and the pursuit of “dual carbon goals”. For local leaders, reducing carbon emissions has now become one of the key achievements they must strive for. In regions where ECP policies are implemented, local governments acting as “self-interested actors” actively engage in ecological civilization construction. They diligently adhere to established objectives and compete with other local governments to gain recognition from the central government, achieve political victories, and obtain political incentives [23]. At this point, local governments invest more effort in ECP projects, mobilize resources to enhance the “innovation compensation” effect, and stimulate regional enterprises to have a greater innovation drive.

However, incentives do not necessarily lead to good results [24]. Empirical evidence indicates that high-ranking local officials in China typically hold their positions for a duration of 3–5 years. In order to achieve quick results, these officials tend to prioritize policies with short-term objectives. In contrast, leadership transitions may occur during the extended policy period in contexts defined by long-term objectives. The political legacy of the departing official is inherited by their successor, which can reduce the motivation of local officials. As a result, local authorities may prefer a strategy that emphasizes short-term incentives.

In this context, we believe that local officials are more inclined to drive policies aimed at reducing carbon emissions when short-term incentives are at play. Conversely, in regions with longer-term incentives, local officials may lack the motivation to drive policy implementation. This results in policies that help enterprises generate an “innovation compensation” but fail to offset “compliance costs”. Consequently, “compliance costs” exceed “innovation compensation”, leading to a reduction in carbon intensity but not an improvement in carbon efficiency.

Hypothesis 3a (H3a).

Compared to long-term incentives, ECP policies with short-term incentives are more effective in reducing carbon intensity and improving carbon efficiency through reasonable incentive mechanisms.

2.3.2. Constraint Mechanism

Throughout the policy implementation process, local governments serving as “agents” entrusted with specific responsibilities often find themselves having to complete multiple tasks assigned by the central government simultaneously. Consequently, when conflicts of interest emerge between the central and local governments, local governments may utilize their discretion, granted by the central government, to adopt different strategies for implementing pilot policies based on the level of alignment of interests and the pressure to execute. This is often referred to as a “differential coping” strategy. In the absence of effective supervision, even though local governments, in their role as “agents”, may actively engage in the application of pilot policies in response to the central government they may lack the sustained motivation to persistently and actively carry out these policy experiments [25]. Consequently, they will not enact stringent environmental regulatory measures, thus failing to impact carbon emissions within their respective regions. To regulate the “differential coping” strategy of local governments, the central government needs to establish certain constraint mechanisms, whether they be in the form of assessment systems or penalty systems, to standardize the behavior of local governments and prevent any potential negligence.

So, it is crucial that the stipulations of these constraint mechanisms are rational. Otherwise, in regions subjected to higher levels of constraint, while the central government’s assessments might deter local governments from inaction to some extent, excessive levels of constraint may induce instances of data concealment by local authorities. Several scholars have explored the phenomenon of firms opting for quantity innovation over quality innovation [26]. The results of these studies indicate that when faced with innovation incentives companies tend to actively increase the number of patent applications. However, this growth primarily focuses on the “quantity” rather than the “quality” of innovation. As a result, there is an increase in non-invention patents, but no significant improvement in technology or product quality is achieved.

When environmental regulatory policies become excessively stringent, local officials may resort to strategic innovation. This response arises from the adverse impact of overly rigorous environmental regulations on the production and operations of local businesses, resulting in heightened operational costs. In their efforts to alleviate these costs and maintain the performance of local governments, local officials may choose innovations that demonstrate superficial compliance with regulations while lacking substantive environmental improvements. Such innovations may encompass compliance-focused pollution control measures, which may not necessarily lead to significant environmental enhancements. This means that local officials may prioritize performance and business operations over genuine efforts to drive sustainable environmental improvements.

In the broader context, it is essential to strike a balance between innovation incentives and regulatory constraints. An overemphasis on either side can result in adverse consequences. Within the framework of China’s political landscape, when constraint intensity is reasonably calibrated, ECP policies can encourage local officials to focus on substantial innovation. This entails implementing significant measures to drive technological innovation and motivating enterprises to attain “innovation compensation” surpassing their “compliance costs”. However, if constraint intensity is excessively high, local officials, acting as intermediaries, may opt for strategic innovation due to the disproportionate effort-to-reward ratio. This choice is made to maintain their political performance and ensure the stable operation of enterprises. In such scenarios, even though local governments, acting as “agents” for the central government, might still actively participate in the application of relevant pilot policies, due to the unreasonable constraint setting they lack the motivation to continue active policy experimentation. This, in turn, hinders the implementation of robust environmental regulatory measures, resulting in the ineffectiveness of both “compliance costs” and “innovation rewards” throughout this process.

Hypothesis 3b (H3b).

As the constraint intensity of ECP policies transitions from weak to strong, local governments tend to favor strategic innovation over substantive innovation.

3. Materials and Methods

3.1. DID Model with Multiple Periods

The ECP policies can be regarded as a quasi-natural experiment in which the selection of policy pilot areas is deliberately controlled, resulting in a degree of artificial selectivity in the grouping of experimental and control units. Given that the ECP policies examined in this study have multiple time points, the traditional Difference-in-Differences (DID) model, which is typically used to assess policy effects at a single time point, is not suitable. Therefore, this study adopted the approach proposed by Beck et al. [27] to construct a DID model with multiple periods in order to evaluate the impact of ECP policies on carbon emissions. The specific formula used is as follows:

$${EI}_{it}={\alpha }_i+{\beta }_i·{Post}_{it}·{Treat}_{it}+\sum {\gamma ·Controls}_{it}+v_{year}+{\mu }_{city}+{\epsilon }_{it}$$

(1)

$${EFF}_{it}={\alpha }_i+{\beta }_i·{Post}_{it}·{Treat}_{it}+\sum {\gamma ·Controls}_{it}+v_{year}+{\mu }_{city}+{\epsilon }_{it}$$

(2)

where EI_it is the carbon intensity in the city i and year t, EFF_it represents the carbon efficiency in the city i and year t. Post_it and Treat_it are time and policy dummy variables. Specifically, if the city i is selected for the ECP Policies, its policy dummy variable Treat_it = 1; otherwise, it is set to 0. Similarly, if the city i is included in the ECP Policies in the year t, the ${Post}_{it}$ and Post_i,t+n are all set to 1; otherwise, Post_it = 0. The interaction term Post_it × Treat_it is the explanatory variable in the model. Furthermore, we included a set of control variables denoted as Controls_it to account for other potential factors that may influence carbon intensity and efficiency. The variables v_year and μ_city represent time-fixed effects and city-fixed effects, respectively. The error term is represented as ε_it.

3.2. Impact Mechanisms

Apart from direct regulatory measures, ECP policies commonly exert influence on carbon emissions through two primary pathways: technological progress and industrial optimization. Drawing upon the methodology suggested by Akerman et al. [28], we incorporated interaction terms into a DID model to discern the distinct effects between the intergroup coefficient and the policy. The formula employed is as follows:

$${EI}_{it}={\alpha }_1+{\beta }_1·{Post}_{it}·{Treat}_{it}+{{\beta }_2·Post}_{it}·{Treat}_{it}·{Effect}_{it}+\sum {\gamma ·Controls}_{it}+v_{year}+{\mu }_{city}+{\epsilon }_{it}$$

(3)

$${EFF}_{it}={\alpha }_1+{\beta }_1·{Post}_{it}·{Treat}_{it}{{+\beta }_2·Post}_{it}·{Treat}_{it}·{Effect}_{it}+\sum {\gamma ·Controls}_{it}+v_{year}+{\mu }_{city}+{\epsilon }_{it}$$

(4)

When examining the impact of technological progress, substitute the technology effect variable Tech_it for Effect_it. When assessing the effect of industrial optimization, replace Effect_it with the industrial effect variable Oppti_it, and then compare the differences between β₁ and β₂ to discern the policy’s influence. The data source is the “China City Statistical Yearbook”.

3.3. Incentive–Constraint Mechanisms

3.3.1. Incentive Mechanisms

China’s current territorial and quantified evaluation system serves as a strong motivator for local officials. This “incentive-oriented” instrument capitalizes on the competitive nature of promotions among local officials, functioning as a mechanism to stimulate their dedication. Importantly, it fosters a spirit of mutual competition among local officials, thereby further incentivizing local governments to intensify their efforts in implementing ECP policies, all in the pursuit of their own self-interest.

We constructed incentive variables based on the important indicators of local officials’ promotion tournament, the incentive virtual variable Int_it and Post_it × Treat_it, to construct a triple interaction term. Based on relevant studies [29], empirical testing is conducted using the DDD model with the following formulas:

$${EI}_{it}={\alpha }_1+{\beta }_1·{Post}_{it}·{Treat}_{it}·{Int}_{it}+\sum {\gamma ·Controls}_{it}+v_{year}+{\mu }_{city}+{\epsilon }_{it}$$

(5)

$${EFF}_{it}={\alpha }_1+{\beta }_1·{Post}_{it}·{Treat}_{it}·{Int}_{it}+\sum {\gamma ·Controls}_{it}+v_{year}+{\mu }_{city}+{\epsilon }_{it}$$

(6)

Regarding the ${Int}_{it}$ variable, we stipulate that if the mayor or party secretary of a city is promoted in the current year, the official promotion index for that year is defined as 1. If both the mayor and the party secretary are promoted, the concept’s official promotion index is defined as 2. The cumulative intensity of official promotion within 12 years is then calculated to determine the official promotion frequency of the city. The data on official changes in Chinese cities were collected manually by the author through the internet.

3.3.2. Constraint Mechanisms

Technological innovation is a critical factor in the effectiveness of ECP policies in promoting the reduction of carbon emissions. In the context of vertical constraints between central and local governments, local governments are likely to use the number of patents as a performance indicator for regional innovation when reporting to the central government in order to serve their self-interest. Therefore, we decomposed the dependent variable into “substantial innovation” and “strategic innovation”, constructing a DID model to examine the constraint intensity of different policies, with the following formulas:

$${TI}_{it}={\alpha }_1+{\beta }_1·{Post}_{it}·{Treat}_{it}+\sum {\gamma ·Controls}_{it}+v_{year}+{\mu }_{city}+{\epsilon }_{it}$$

(7)

$${SI}_{it}={\alpha }_1+{\beta }_1·{Post}_{it}·{Treat}_{it}+\sum {\gamma ·Controls}_{it}+v_{year}+{\mu }_{city}+{\epsilon }_{it}$$

(8)

TI_it represents “substantive innovation”, using the number of green patent inventions as a proxy variable while SI_it represents “strategic innovation”, using the number of utility model patents and design patents as proxy variables. The patent data comes from the China State Intellectual Property Office.

3.4. Dependent Variable

To measure the carbon emission reduction effects, this paper assesses the carbon emissions of various prefecture-level cities in China from two dimensions: carbon intensity and carbon efficiency.

Carbon intensity (EI_it) refers to the carbon dioxide emissions per unit of GDP, reflecting the relative relationship between economic growth and carbon emissions. The calculation of carbon dioxide emissions involves aggregating the carbon emissions resulting from coal gas and liquefied petroleum gas consumption, electricity usage, transportation activities, and heat energy consumption within each city. The carbon emissions resulting from the consumption of coal gas and liquefied petroleum gas are calculated using conversion factors provided by the Intergovernmental Panel on Climate Change (IPCC) in 2006. The carbon emissions resulting from electricity consumption are calculated using the emission factors of the regional power grid [30]. The carbon emissions generated by transportation are calculated using the passenger and freight volumes of different transportation modes in the city [31]. Lastly, the carbon emissions resulting from thermal energy consumption are calculated by considering the amount of raw coal consumed by the city. All carbon emissions are then added together to obtain the total carbon emissions.

Carbon efficiency (EFF_it) is measured using the slacks-based model (SBM) with non-desirable outputs proposed by Tone [32]. In this model, Chinese cities are considered as distinct decision-making units, each having three vectors: inputs, expected outputs, and unexpected outputs. Specifically, the input vector can be denoted as X ∈ R^m, the expected output vector as Y ∈ R^q, and the non-expected output vector as B ∈ R^p. The input matrix, expected output matrix, and non-expected output matrix can be defined as follows: X = [x₁,x₂…,x_n] = R^(m×n), Y = [y₁,y₂…,y_n] = Y^(q×n), B = [b₁,b₂…,b_n] = R^(p×n).

Assuming X > 0, Y > 0, B > 0, the production possibility set is P = {(x,y,b)|x ≥ X_λ, y ≥ Y_λ, b ≥ B_λ, λ > 0}, λ is weight vector, ρ* is the objective function.

$$\rho \mathrm{*}=min\frac{1-\frac{1}{m}{\sum }_{i=1}^m\frac{{s_i}^{-}}{x_{ik}}}{1+\frac{1}{k_1+k_2}\left({\sum }_{r=1}^{k_1}\frac{{s_r}^{+}}{y_{rk}}+{\sum }_{t=1}^{k_2}\frac{{s_t}^{-}}{b_{tk}}\right)}$$

(9)

$$\begin{array}{c} s.t. {\mathrm{x}}_{ik}|={\displaystyle \sum _{j=1}^n}{\lambda }_j{\mathrm{x}}_{ij}+{s_i}^{-} i=1,2,\dots ,m;\\ {\mathrm{y}}_{rk}={\displaystyle \sum _{j=1}^n}{\lambda }_j{\mathrm{y}}_{rj}+{s_r}^{+} r=\mathrm{1,2},\dots ,q;\\ {\mathrm{b}}_{tk}={\displaystyle \sum _{j=1}^n}{\lambda }_j{\mathrm{b}}_{tj}+{s_t}^{-} t=\mathrm{1,2},\dots ,p;\\ {\lambda }_j\ge 0, {s_i}^{-}\ge 0, {s_r}^{+}\ge 0, {s_t}^{-}\ge 0\end{array}$$

s represents the slack variables for input vectors, expected output vectors, and non-expected output vectors, while λ stands for the weight vector. When 0 < ρ* < 1, it indicates the presence of some efficiency loss, which means that carbon efficiency can be improved through enhancements in input and output units. When the objective function ρ* ≥ 1, it signifies high input–output efficiency for the region, with higher ρ* values indicating higher carbon efficiency in that area. We use year-end employment figures as human input, the capital stock formed by fixed asset investments based on the 2006 as capital input, urban electricity generation as energy input, GDP as economic output, and carbon dioxide emissions as non-expected output to measure carbon efficiency in Chinese cities. The data used in the calculation all comes from the “China City Statistical Yearbook” and “China Energy Statistical Yearbook”.

3.5. Control Variables

In view of the available statistical data, we conducted empirical testing using a panel dataset covering 224 cities in China from 2009 to 2020. Acknowledging that other urban characteristics might potentially impact carbon emissions, we included the following control variables: urbanization level, denoted as the ratio of urban population to the total regional population; financial level, quantified as the ratio of local loans to GDP; industrial level, captured by the share of value added by the secondary industry in the total output; infrastructure development level, measured as the proportion of road mileage to urban area for each prefecture-level city; fiscal decentralization level, represented by the ratio of city fiscal revenue to GDP. Data sources include the “China City Statistical Yearbook” and the National Bureau of Statistics of China.

4. Results

4.1. Main Results

The results in

Table 2. Main Regression.

	EI	EFF	EI	EFF	EI	EFF
A1	−2.088 **	0.132 ***
	(0.818)	(0.026)
A2			−0.605 *	−0.009
			(0.333)	(0.011)
A1 × A2					−2.620 **	0.182 ***
					(1.315)	(0.042)
Urban	2.291 **	−0.003	2.295 **	0.000	2.321 **	−0.005
	(0.955)	(0.031)	(0.956)	(0.031)	(0.956)	(0.031)
Finance	1.275 ***	−0.040 ***	1.271 ***	−0.039 ***	1.272 ***	−0.039 ***
	(0.247)	(0.008)	(0.247)	(0.008)	(0.247)	(0.008)
Industry	1.532 ***	−0.049 ***	1.590 ***	−0.051 ***	1.559 ***	−0.050 ***
	(0.321)	(0.010)	(0.321)	(0.010)	(0.321)	(0.010)
Facility	−0.134 ***	0.005 ***	−0.125 ***	0.004 ***	−0.140 ***	0.005 ***
	(0.044)	(0.001)	(0.044)	(0.001)	(0.044)	(0.001)
Fiscal	21.632 ***	−1.144 ***	22.003 ***	−1.142 ***	22.042 ***	−1.173 ***
	(6.291)	(0.202)	(6.298)	(0.203)	(6.297)	(0.203)
Constant	−18.498 ***	1.250 ***	−19.304 ***	1.281 ***	−18.878 ***	1.273 ***
	(4.428)	(0.142)	(4.428)	(0.143)	(4.426)	(0.142)
Observations	2688	2688	2688	2688	2688	2688
Adjust R2	0.629	0.680	0.628	0.677	0.628	0.679
City fixed effects	Yes	Yes	Yes	Yes	Yes	Yes
Year fixed effects	Yes	Yes	Yes	Yes	Yes	Yes

Notes: *, **, and *** mean significant at the 10%, 5%, and 1% level, respectively.

show that, irrespective of the implementation of one or both policies, cities within the pilot scope have passed the test. This implies that the implementation of the ECP policies in these regions has effectively led to a reduction in carbon intensity within the respective areas. This substantiates that the ECP policies currently enforced in China have significantly promoted carbon emission reduction and positively influenced the ecological environment development of the energy industry. Furthermore, as one progresses from Policy A2 to Policy A1 and subsequently to regions where both policies are simultaneously implemented the policy intensity escalates, yielding a corresponding increase in its impact on carbon intensity and carbon efficiency. Policy A1 demonstrates effects on carbon intensity and carbon efficiency, registering values of −2.088 and 0.132, which are notably higher than those of Policy A2. In cities that implement both policies concurrently, the effects of the pilot policy on carbon intensity and carbon efficiency are −2.620 and 0.182, surpassing the outcomes observed in cities implementing a single policy. This substantiates Hypothesis 1 that ECP policy can reduce the carbon intensity of pilot cities and improve carbon efficiency in the energy industry. Additionally, as the policy intensity of ECP policies gradually increases, the policy effect also improves.

In the experiment on control variables, we find that financialization, industrialization, and the enhancement of fiscal decentralization all had a significant impact in elevating carbon intensity and reducing carbon efficiency. It is noteworthy that urbanization was the sole exception, as it increased carbon intensity but did not reduce carbon efficiency. These findings align with logical reasoning and prior experimental knowledge. Interestingly, among the controlled variables, the improvement in infrastructure construction had a substantial effect on decreasing carbon intensity and enhancing carbon efficiency. While the influence of improved infrastructure development on carbon reduction is relatively modest when compared to the ECP pilot policy, it substantiates China’s capacity to mitigate carbon emissions by its notable externalities.

4.2. Placebo Test

Although we controlled for urban variables in the experiment, it was imperative to enhance the robustness of our regression results. To achieve this, we randomly selected an equivalent number of cities from all sample cities as the control group for placebo tests [33]. Employing a random sampling approach, we generated 500 sets of placebo variables. The resulting kernel density and coefficient distributions were then depicted in figures, allowing for a comparison with the original findings and the presentation of the placebo test outcomes.

Figure 1 illustrates the results of the placebo tests for Policy A1 and Policy A2. Notably, the sampled outcomes of carbon intensity and carbon efficiency for Policy A1 significantly deviate from the original results, substantiating that our experimental findings do not exhibit a placebo effect. While the effect of Policy A2 on carbon intensity also passes the test, the results for carbon efficiency, due to their lack of significance within the policy itself, do not exhibit significant differences compared to the placebo test. These test results are consistent with the main regression findings, demonstrating the robustness of the experiment.

4.3. Impact Mechanisms Results

Table 3. How ECP policies affect carbon emissions reductions by industrial structure optimization.

	EI	EFF	EI	EFF
A1	36.092 **	−1.253 **
	(16.210)	(0.521)
A1 × Ins	−5.595 **	0.203 ***
	(2.372)	(0.076)
A2			15.959 ***	−0.387 **
			(4.900)	(0.158)
A2 × Ins			−2.511 ***	0.057 **
			(0.741)	(0.024)
Constant	−18.451 ***	1.248 ***	−18.886 ***	1.272 ***
	(4.424)	(0.142)	(4.420)	(0.143)
Observations	2688	2688	2688	2688
Adjust R2	0.630	0.681	0.630	0.678
Control	Yes	Yes	Yes	Yes
City fixed effects	Yes	Yes	Yes	Yes
Year fixed effects	Yes	Yes	Yes	Yes

Notes: ** and *** mean significant at the 5%, and 1% level respectively.

and

Table 4. How ECP policies affect carbon emissions reductions by technological progress.

	EI	EFF	EI	EFF
A1	5.053 *	−0.149 *
	(2.620)	(0.084)
A1 × Tech	−0.998 ***	0.039 ***
	(0.348)	(0.011)
A2			2.480 ***	−0.193 ***
			(0.931)	(0.030)
A2 × Tech			−0.524 ***	0.031 ***
			(0.148)	(0.005)
Constant	−18.506 ***	1.249 ***	−18.603 ***	1.239 ***
	(4.427)	(0.142)	(4.426)	(0.142)
Observations	2678	2678	2678	2678
Adjust R2	0.631	0.681	0.631	0.682
Control	Yes	Yes	Yes	Yes
City fixed effects	Yes	Yes	Yes	Yes
Year fixed effects	Yes	Yes	Yes	Yes

Notes: * and *** mean significant at the 10% and 1% level respectively.

present how ECP policies influence carbon reduction in the urban green energy industry. In Table 3, we find after introducing the interaction terms in the DID model that the estimated value β₂ of the interaction term is significantly smaller than the estimated value β₁ of ECP policies for carbon intensity. Similar results manifest in the context of carbon efficiency. These findings substantiate the implications of energy industrial structure optimization, thus affirming Hypothesis 2a. ECP policies effectively reduce carbon intensity and enhance carbon efficiency by stimulating energy industrial structure optimization. The results in Table 4 are consistent with those in Table 3. The incorporation of interaction terms leads to a notable decrease in carbon intensity and a significant improvement in carbon efficiency, thus providing empirical support for Hypothesis 2b. ECP policies can reduce carbon intensity and enhance carbon efficiency by promoting technological progress.

We also found that the changes resulting from the inclusion of energy industrial structure optimization variables exceeded those induced by technological progress variables. This finding suggests that the effect of energy industrial structure optimization has a more significant influence than green technological progress. As previously mentioned, whether from the Pollution Haven Theory or the Environmental Regulation Theory perspective, the optimization of industrial structure has a positive impact on carbon reduction objectives. In contrast, technological progress exhibits a dual effect, both promoting carbon reduction through “innovation compensation” and lowering carbon efficiency due to “compliance costs”. Although the experiment didn’t conclusively affirm this proposition, it indirectly lends credence to its plausibility.

4.4. Incentive-Constraint Mechanism

Local governments have historically played a significant role in fostering regional economic and social progress. Within the context of the specific central–local interaction mechanism, it is important to highlight the dual roles played by the central government. On one hand, the central government employs constraint mechanisms, which exert a vertical constraint effect on local governments. This serves as a vital means to mitigate potential moral hazards that may arise among local governments. On the other hand, the central government strategically employs incentive mechanisms to motivate local governments, acting as self-interested actors, to actively pursue elevated economic and political benefits.

4.4.1. Incentive Mechanism

The effectiveness of “horizontal competition” stems from the capacity of local governments, operating within this framework, to optimize their interests through the exercise of administrative authority. Consequently, for the highest-ranking executive officials within local governments, the strength of incentives, reflecting the alignment of interests among these officials, takes the form of a “promotion” system within the Chinese context. In this system, when a local government official demonstrates outstanding abilities in the region under their jurisdiction, they frequently experience expedited career advancement. This dynamic underscores one of the merits of the promotion tournament, as it facilitates the rapid elevation of competent officials.

Table 5. Impact of policy incentive mechanism.

	(1)	(2)	(3)	(4)
	EI	EFF	EI	EFF
A1 × Int	−3.974 ***	0.244 ***
	(1.418)	(0.046)
A2 × Int			−0.982 *	−0.004
			(0.536)	(0.017)
Constant	−18.559 ***	1.255 ***	−19.289 ***	1.284 ***
	(4.425)	(0.142)	(4.427)	(0.143)
Observations	2688	2688	2688	2688
Adjust R2	0.629	0.680	0.628	0.677
Control	Yes	Yes	Yes	Yes
City fixed effects	Yes	Yes	Yes	Yes
Year fixed effects	Yes	Yes	Yes	Yes

Notes: * and *** mean significant at the 10% and 1% level, respectively.

reveals the examination results of the incentive mechanism. Regions that have implemented ECP Policies, whether it is Policy A1 or Policy A2, exhibit significant correlation between the level of official promotions and the effectiveness of the policy. This phenomenon indicates the presence of an official promotion tournament within China’s pilot policies. These ECP policies can promote carbon reduction through the mechanisms related to official promotions. In the process of implementing ECP policies, when a region has a higher intensity of official promotions, its local leaders will exert more effort to enhance the policy effects of the ecological civilization pilot, thereby positioning themselves more favorably in the official promotion tournament.

In this study, we also observed that, in comparison to areas which implemented Policy A2, Policy A1 demonstrated a more pronounced promotion of carbon reduction and enhancement of carbon efficiency. This observation indicates that Policy A1 as a “green honor”, can effectively motivate local officials to elevate their ecological civilization development efforts. Such endeavors result in decreased carbon emissions and improved carbon efficiency. Contrarily, Policy A2, which offers long-term incentives, failed to produce a similar effect on carbon efficiency through the official promotion tournament. These findings provide empirical support for Hypothesis 3a, indicating that ECP policies with short-term incentives are more effective in reducing carbon intensity and enhancing carbon efficiency when employing a well-structured incentive mechanism.

4.4.2. Constraint Mechanism

Table 6. Impact of policy constraint mechanism.

	Substantive Innovation	Strategic Innovation	Substantive Innovation	Strategic Innovation
A1	0.230 *	−0.282 **
	(0.131)	(0.112)
A2			−0.149 ***	−0.151 ***
			(0.053)	(0.045)
Constant	2.922 ***	1.003 *	2.922 ***	0.866
	(0.707)	(0.605)	(0.706)	(0.604)
Observations	2688	2688	2688	2688
Adjust R2	0.901	0.925	0.901	0.925
Control	Yes	Yes	Yes	Yes
City fixed effects	Yes	Yes	Yes	Yes
Year fixed effects	Yes	Yes	Yes	Yes

Notes: *, **, and *** mean significant at the 10%, 5%, and 1% level respectively.

presents the examination results of the constraint mechanism. Policy A2, which features an elimination system, is expected to have a stricter assessment compared to Policy A1. However, in practical policy implementation, Policy A1 follows fixed assessment criteria, while Policy A2 allows self-declared criteria. To facilitate their assessment approval, local officials tend to opt for easily achievable indicators. This tendency is particularly pronounced in terms of innovation. We find that both ECP policies have a significantly negative impact on strategic innovation. The implementation of these policies effectively reduces the level of strategic innovation. Additionally, the impact of Policy A2 on substantive innovation is notably lower than that of Policy A1. Policy A1 exhibits a significantly more substantial influence in diminishing strategic innovation when contrasted with Policy A2. This observation confirms Hypothesis 3b, suggesting that as the strength of ECP policies shifts from weak to strong, local governments tend to lean more towards strategic innovation rather than substantive innovation. This indicates that the assessment mechanism set in ECP policies makes local governments refrain from using strategic innovation to bypass central governments. The increased constraint does indeed somewhat reduce the behavior of local governments pursuing innovation quantity over innovation quality for political gains. Importantly, the stringency of the assessment is not the sole factor; the reasonable design of assessment criteria has the potential to more effectively promote substantive innovation while diminishing strategic innovation.

Combined with the results in Table 5 and Table 6, it is evident that the incentive–constraint mechanism formulated by Policy A1 is more reasonable. Both its incentive and constraint mechanisms contribute to the enhancement of regional ecological civilization construction, thereby promoting the reduction of carbon emissions and the enhancement of carbon efficiency within the region. In contrast, Policy A2 leads to a concurrent decline in both strategic and substantial innovation levels. In regions where ECP policies are implemented with heightened assessment intensity, the continuous rise in “compliance costs” results in a crowding-out effect, leading to a decrease in the level of “substantial innovation” among enterprises. Concurrently, due to the formidable constraint capacity of the central government, regions that implement ECP policies witness a decrease in the level of “strategic innovation” and it is evident that the incentive–constraint mechanism formulated by Policy A1 is more reasonable. Both its incentive and constraint mechanisms contribute to the enhancement of regional ecological civilization construction, thereby promoting the reduction of carbon emissions and the enhancement of carbon efficiency within the region. In contrast, Policy A2, characterized by higher assessment intensity, leads to a concurrent decline in both strategic and substantial innovation levels. In regions where ECP policies are implemented with heightened assessment intensity, the continuous rise in “compliance costs” results in a crowding-out effect, leading to a decrease in the level of “substantial innovation” among enterprises. Concurrently, due to the formidable constraint capacity of the central government, regions that implement ECP policies witness a decrease in the level of “strategic innovation”.

5. Conclusions

In this study, we have empirically verified Hypotheses 1 through 3, leading us to the unequivocal conclusion that China’s current ECP policies are highly effective in reducing regional carbon emissions. This reduction is principally accomplished through the optimization of industrial structures and the advancement of technological capabilities. Additionally, the Chinese incentive–constraint mechanism plays a pivotal role in this endeavor. Short-term incentives and well-defined assessment standards serve as motivational levers, prompting active participation among local officials in ecological civilization development. These findings not only provide empirical support for the theoretical foundation of ECP policies but also present practical strategies for establishing administrative frameworks dedicated to fostering carbon emission reduction.

Furthermore, our study highlights the evolving role of the official promotion competition in China’s governance framework. It no longer solely prioritizes GDP growth but has progressively encompassed ecological civilization goals. This transformation endows it with substantial influence in incentivizing local government officials to contribute proactively to environmental and ecological initiatives. Our research also uncovers a positive correlation between the comprehensiveness of evaluation criteria within ECP policies and their actual policy impacts. In contrast, permitting local governments to independently formulate assessment standards, while considering local contexts to some extent, does not inherently encourage substantial innovation at the grassroots level. This underscores the critical importance of judicious central government oversight applying reasonable constraints on their subordinate counterparts, thereby fostering greater dedication to ecological civilization construction and ultimately enhancing policy effectiveness.

Based on these findings, we offer the following three recommendations for the central government:

Global Adoption of ECP Policies for Carbon Emission Reduction. It is recommended that governments worldwide consider implementing policies similar to China’s ECP policies. These policies have demonstrated their effectiveness in reducing regional carbon emissions. By adopting ECP-like initiatives, governments can advance their own carbon emission reduction efforts. Emulating successful models and adapting them to local contexts can provide a structured framework for addressing carbon reduction and environmental sustainability at the national level.

Enhanced Guidance for the Energy Industry and Promoting Technological Innovation. To expedite carbon emission reduction and bolster ecological objectives, governments should provide heightened guidance to the energy industry. Policymakers can encourage technological innovation within the sector, emphasizing the development of environmentally friendly and efficient technologies. This approach not only supports carbon reduction but also fuels economic growth by fostering technological advancements that align with ecological and sustainability goals.

Tailored Incentive and Constraint Strategies in ECP Policy Design. When crafting ECP policies, policymakers should take into account the incentive and constraint effects inherent in the policy design. It is essential to customize these strategies based on the specific context and conditions of each country. This tailored approach ensures that the incentive and constraint mechanisms are well-suited to the unique circumstances of each nation. By doing so, governments can maximize the effectiveness of their ecological policies and motivate active participation among stakeholders.

However, it is important to acknowledge the limitations of this research. We focused primarily on China and did not consider the effects of environmental development policies in other nations. In future research, we plan to integrate data from other countries to broaden our understanding of the efficacy of environmental policies on a global scale. Additionally, while we have conducted robustness tests to address endogeneity, we aim to explore more rigorous methods for handling endogeneity and reducing potential estimation biases in later stages. This is in line with the advancement of quantitative research techniques and will contribute to a more comprehensive understanding of the subject.

In summary, our study contributes to the understanding of the impact of ECP policies on carbon emissions in the urban green energy industry. The policy recommendations presented here aim to further enhance environmental policies and carbon reduction efforts in China, while recognizing the need for more extensive research and methodological advancements in the field.