Can New Energy Become a Breakthrough for Economic Development—Based on Clean Development Mechanism Projects in Less Developed Coastal Cities

Abstract

Coastal cities have the natural resource endowment and location advantages to develop new energy. However, heterogeneity in the economic development of China’s coastal cities has led to differences in the outcomes of environmental regulatory policies and related programs. To elucidate the difference, this paper obtained 5074 clean development mechanism (CDM) projects, which serves as a key instrument of the Kyoto Protocol designed to assist developing countries in achieving sustainable development through project-based emissions reductions and conducted a causal identification through quasi-experiment. And DID as well as DDD models are applied to identify the CDM effects on cities’ economic development. Main findings are: (1) Through the DID regression, this paper finds that the development of CDM projects have promoted the development of the city’s economy and lead to the upgrading of cities’ industries. (2) The promoting effects in economic development and employment are more prominent in coastal cities with high levels of economic development. (3) CDM can better facilitate economic development and employment in less developed coastal areas when implemented in conjunction with economic promoting policies. By applying quasi-experimental methods, including DID and DDD models, the research introduces a novel approach to assess the causal effects of CDM projects on city economies, offering fresh insights into sustainable development policies.

1. Introduction

1.1. Status of Less Developed Coastal Cities

Except for Hong Kong, Macao, Taiwan and Sansha, there are 54 coastal cities at the prefecture level and above in China. Statistics for 2018 show that 19 of these coastal cities have a GDP per capita less than the national average level, which is 66,000 CNY. And 33 coastal cities have a GDP per capita less than the average level in 11 eastern provinces and cities, which is 87,000 CNY, falling into the less developed areas of the eastern region. These cities are called less developed coastal cities. There are 12 least developed coastal cities with per capita GDP less than 50,000 yuan, of which Guangdong Province accounted for six. Shanwei is the coastal city with the lowest GDP per capita, with a GDP per capita of only 30,800 CNY, which is 500 CNY lower than that of Gansu, the province with the lowest GDP per capita in the country. The development of these less developed coastal cities has been a failure relative to the country’s 40-year-long high economic growth.

Underdeveloped coastal cities face unique challenges in terms of energy access due to their particular geographic location. These cities often lack adequate energy infrastructure, which limits economic development and the quality of life of their residents [1]. Unequal access to energy not only affects the daily lives of residents but can also lead to uneven regional development. Many underdeveloped coastal cities are located in areas where energy resources are scarce, which limits the local energy supply capacity. At the same time, due to geographical constraints, these cities are highly dependent on external inputs for energy supply, which increases energy costs [2]. Compared with developed regions, underdeveloped coastal cities have deficiencies in energy policy support and investment, which directly affects the application and diffusion of new energy technologies. In addition, the potential impacts of climate change cannot be ignored. Climate change facing coastal areas may cause serious damage to energy infrastructure. For example, sea level rise threatens coastal energy facilities, while extreme weather events may interrupt energy supply [3].

Most of the less developed coastal cities have good harbor transportation conditions and are also the areas with the most abundant supply of diversified fruits and vegetables in China due to the high precipitation. And many cities are also estuaries of navigable inland rivers. Therefore, it can be said that the resources and location conditions of these cities are very superior, which is one of the best development conditions in China. Many less developed coastal cities, in the early reform and opening up, also once became star cities, such as Shantou, Wenzhou. Shantou is one of the four special economic zones, and Wenzhou was once known for a private economic model called the Wenzhou model. In terms of the development period, the development gap between the less developed coastal cities and the developed coastal cities has gradually widened since this century.

China’s coastal 54 prefecture-level cities account for less than 22% of the country’s population, while more than 60% of the world’s population live in the coastal areas within 60 km [4]. The U.S. coastal zone, which runs 80 km deep, is home to 57% of the population. China is also the country with the highest population density inland, more than 80% of the world’s non-coastal offshore and non-navigable megalopolis of more than 3 million people are located in China [5]. However, roughly since 2004, China’s inland provinces and regions have been less efficient than coastal provinces and regions in terms of macro investment output, and the gap continues to widen. In 2018, the macro investment output rate in the central and western inland was only 61.6% of that in the eastern coastal provinces and regions. This may mean that the optimization of the country’s economic geography may also require large-scale inland-coastal population migration, and the coastal areas need to open new spaces for population and economic agglomeration to take over. Therefore, focusing this study on the development breakthrough of less developed coastal cities will certainly provide strong theoretical support and practical guidance for the strategy and policy formulation of all levels of government on the development breakthrough of less developed coastal cities.

1.2. The Advantages of Developing New Energy in Less Developed Coastal Cities

China’s coastal areas has a unique location condition in the development of new energy. First of all, from the wind power aspect. In recent years, as China’s onshore wind power construction technology has become increasingly mature, and offshore wind power resources are more extensive, the national wind power development policy gradually tilted to offshore power generation [6]. Compared to onshore wind farms, the advantages of offshore wind farms are mainly that they do not occupy land resources, are largely unaffected by topography, have higher wind speeds, have larger wind turbine unit capacities, and have higher annual utilization hours. In 2020, China’s new wind power installed on the grid reached 71.67 million kilowatts, of which 3.06 million kilowatts of new offshore wind power installed. By the end of 2020, the cumulative installed capacity of offshore wind power in China is about 9 million kilowatts. According to “Research on China’s ‘14th Five-Year’ electric power development planning”, China will mainly develop offshore wind power in Guangdong, Jiangsu, Fujian, Zhejiang, Shandong, Liaoning, and Guangxi coastal areas, focusing on the development of seven large offshore wind power bases. It is expected that in 2035, 2050, the total installed capacity of large bases will reach 71 million, 132 million kilowatts, respectively.

Secondly, as shown in Figure 1, from the perspective of light resources, the overall light conditions in the eastern part of China are average. However, some coastal areas or cities generally have longer sunshine hours because clouds cannot be gathered, and have light advantages compared with neighboring inland cities.

Finally, from the perspective of nuclear power development, a 1 million kilowatts nuclear power plant will become “sweaty” when it is in operation. Therefore, it requires a large amount of cooling water, about 11 to 12 cubic meters of cooling water per second, and it is usually cooled by using cooling towers in a closed loop. If the turbine is cooled by the condenser at one time, the cooling water requirement must not be less than 50 cubic meters per second. This is 50% higher than a large thermal power plant. Therefore, when selecting sites for nuclear power plants, the focus is often on coastal areas where water is abundant. This is because the vast sea and river waters provide sufficient water for cooling nuclear power plant equipment. Therefore, China’s current nuclear power plant projects are all chosen to be built in coastal cities.

Coastal cities have the natural resource endowment and location advantages to develop new energy, so can new energy become a breakthrough in the economic development of less developed coastal cities? This is the question that this study focuses on.

2. Literature Review

Scholars have conducted a lot of research on CDM, and they have different views on the impact of CDM. On the one hand, CDM can bring positive impacts on the economic, social, and environmental aspects of sustainable development. First, in an analysis of 744 CDM projects, Olsen and Fenhann show that CDM can promote employment, foster economic growth, improve air quality, help access to energy, and create citizen welfare [7]. Fernández et al. argue that CDM contributes more to sustainable development than other types of projects in terms of job creation [8]. Chirambo demonstrates that CDM can promote energy access in underdeveloped regions [9]. In addition, CDM has the potential to contribute to carbon reduction [10,11]. Cui et al. argue that CDM has a positive impact on companies’ innovation in renewable energy use and energy efficiency, and that this positive impact is particularly evident in invention patents [12]. There are also many scholars who believe that CDM programs have a positive impact on technology transfer [12,13,14], and in particular, Watson et al. argue that CDM has an important impact on technology development in China [15].

On the other hand, some scholars argue that the impact of CDM on sustainable development is neutral or even negative [16]. Jaraitė et al. confirmed that CDM projects did not reduce CO₂ emissions, as the CO₂ emissions produced by companies enrolled in CDM projects increased [17]. Zhang and Wang demonstrated that CDM had no positive impact on reducing SO₂ emissions [18]. Zhang et al. demonstrated that the total factor energy efficiency and carbon emission performance of the host country can make CDM negatively impact their energy efficiency [19]. In a sample estimation of 114 CDM projects, Crowe found that 74% of the projects could not contribute to poverty eradication and 16% had only a weak impact on poverty reduction [20]. Pécastaing et al. confirmed that CDM has a weak effect on household consumption expenditure, but no statistically significant effect on employment and poverty reduction [21]. In addition, Sirohi points out that the impact of CDM projects on socioeconomic development in India is ambiguous and argues that CDM should be a win-win strategy [22].

The economic implications of CDM projects have been a particular focus, with studies suggesting that they can significantly boost local economies, particularly in the long term, by optimizing industrial structures, increasing capital stock, and raising fiscal revenue [23]. However, the economic benefits of CDM projects are not evenly distributed, with larger emerging economies tending to gain more than smaller or less developed countries [24]. In the context of coastal cities, which are often less developed and more vulnerable to climate change, the development of new energy is seen as a critical factor in economic development. It can lead to diversification of energy sources, reduction in energy costs, enhancement of energy security, and stimulation of new industries and jobs related to renewable energy. Moreover, the integration of renewable energy into coastal cities’ infrastructure can also contribute to climate change adaptation and disaster risk reduction, which are essential for sustainable development [1].

Despite these potential benefits, the effectiveness of CDM in promoting new energy development in less developed coastal cities remains a subject of debate. Therefore, this study aims to use data from CDM projects in less developed coastal cities to assess whether new energy can contribute significantly to cities’ economic development.

3. Research Design

3.1. Definition of Less Developed Coastal Cities

Except for Hong Kong, Macao, Taiwan, and Sansha, there are 54 coastal cities at the prefecture level and above in China, so the first issue that this study directly faces is the definition of less developed cities. This study intends to use macroeconomic indicators of prefecture-level cities such as GDP per capita, employment ratio, industrial structure, fiscal revenue, etc. to scientifically construct development indicators of cities. Prefectural cities with comprehensive indicators below the average level of national cities are defined as less developed coastal cities.

The Kyoto Protocol, which entered into force in February 2005, is the world’s first legally binding international environmental agreement, which aims to reduce the level of greenhouse gases in the atmosphere through close cooperation between the international community and the protection of the environment [25]. The Kyoto Protocol specifies the greenhouse gas reduction tasks and sets out the operational mechanisms to achieve these reduction targets [26]. The Clean Development Mechanism is one of the three flexible compliance mechanisms stipulated in the Kyoto Protocol; it allows investors from developed countries to implement emission reduction projects in developing countries that contribute to the sustainable development of developing countries [27]. CDM is a win-win mechanism that helps developed countries meet some of their greenhouse gas emission reduction obligations while helping developing countries achieve sustainable development [28]. Since the first CDM project in China was successfully registered with the United Nations CDM Executive Board on 26 June 2005, CDM projects have been rapidly developed in China within a short period of four years. As of 23 August 2016, China ranks first in the world in the number of successfully registered CDM projects, the expected annual emission reductions generated by the projects and the certified emission reductions issued by EB. This not only helps developed countries achieve their greenhouse gas emission reduction targets at a lower economic cost, but also brings in a large amount of capital for Chinese enterprises and social development, introduces advanced ideas and technologies, and has significant social benefits.

China’s first CDM project was approved by the National Development and Reform Commission on 10 March 2005. As of 23 August 2016, 5074 CDM projects have been approved domestically in China, with the expected annual emission reduction of the projects reaching 782 million tons of carbon dioxide equivalent. The projects can be divided into energy saving and energy efficiency category, new and renewable energy category, and methane recycling, waste-to-energy incineration, N₂O decomposition and elimination, afforestation and reforestation according to the type of emission reduction [29]. The details are shown in

Table 1. CDM projects approved by National Development and Reform Commission.

Type of Emission Reduction	Annual Emission Reduction (Million Tons of CO2)	Emission Reduction Ratio	Number of Projects
new and renewable energy	45,940	58.74%	3733
energy saving and energy efficiency	9716	12.42%	623
methane recycling	8212	10.50%	476
fuel substitution	2833	3.62%	51
N2Odecomposition and elimination	2818	3.60%	43
HFC-23 decomposition	6680	8.54%	11
afforestation and reforestation	16	0.02%	5
waste-to-energy incineration	823	1.05%	54
others	1167	1.49%	78
total emission reduction	78,205	100%	5074

.

From the regional dimension, there are a total of 1361 CDM projects in coastal provinces. As shown in Figure 2, the CDM projects in 11 coastal provinces and autonomous regions account for 26.8% of all CDM projects in China.

3.2. Data

First of all, this study obtains all 5074 CDM projects approved by the China CDM website (https://www.cdmfund.org/jjnb.html, accessed on 14 October 2024) up to 23 August 2016 using R 4.4.0 software and crawler technology. The project information includes the specific type, name, completion time and emission reduction of the projects, and the project type information is used to filter the new energy projects.

Second, using the project name and the backend API interface of the Gaode Map, determine the latitude and longitude of the specific location to which the project belongs and the project latitude and longitude information used to determine the city to which the project belongs. Then, screen out new energy projects in less developed coastal cities and other cities.

Finally, the city statistical yearbook is used to integrate city-level macro data, including GDP, GDP per capita, the number of cities’ employees, industrial structure, the number of populations, the paved road area, the average wage of employees, the amount of actual foreign investment, retail sales of social consumer goods, fixed asset investment. These data are used to construct the city-level panel data needed for the study.

As shown in Figure 3, the distribution of new energy CDM projects in China is mainly concentrated in areas rich in wind and solar energy resources. In addition, coastal cities are relatively densely populated with CDM projects. New energy CDM projects show a continuous distribution trend along the coastline.

3.3. Method

This study investigates the impact of new energy projects on the economic development of cities. First, this study needs to observe the changes of economic indicators of cities before and after the project to examine whether there is an impact of new energy projects on the long-term economic development of cities. And in the field of “quasi-natural experiments”, DID models are widely used in policy evaluation. Second, in this study, when observing the changes of economic indicators of cities before and after the project, a comparison study between the experimental and control groups is needed. And PSM is applicable to “non-random data”, which can well solve the problems of sample selection bias and heterogeneity. Therefore, the PSM-DID model is chosen to study the impact of new energy projects on cities’ economic development.

3.3.1. Propensity Score Matching (PSM)

Propensity Score Matching is a statistical method that aims to reduce the problem of sample selection bias by matching methods. In observational studies, there is often a degree of bias in the data, making the data for the control and experimental groups too different, losing some comparability. PSM can eliminate individuals who deviate too much from the overall data of the sample, alleviate the effect of data bias, and make the control and experimental groups tend to be homogeneous and comparable. Specifically, PSM can downscale multiple dimensions of information and calculate a propensity score for everyone, which reflects the degree of similarity between individuals, and if the scores are closer, the samples are more comparable. Therefore, the experimental and control groups of the sample are comparable when each less developed coastal city in the experimental group that has built a large new energy project is matched to some neighboring city that does not have a new energy project.

PSM requires two assumptions to be met, one is the common support hypothesis, the other is the conditional independent distribution hypothesis. The common support hypothesis is that each experimental group and the control group may have a match in the sample, thus ensuring that the experimental group can find a match with the control group. The conditional independent distribution assumption means that the matching variables are random in their distribution after control.

In terms of variable selection for propensity score matching, this study refers to the research method of Du and Takeuchi [30]. First, in order to estimate the propensity score, this study draws on Du and Takeuchi [30]’s practice and selects five variables: city GDP, population, industrial structure, electricity consumption, and fiscal revenue. These covariates are chosen because CDM projects undergo an exhaustive review of project additionality before registration. The additionality of a project is closely related to the characteristics of different regions, especially in terms of economic growth potential and power generation inputs [28]. Second, the study uses estimated propensity scores for year-to-year matching, using a matching methodology of nearest-neighbor 1-to-3 matching.

3.3.2. Difference-in-Differences Model (DID)

The DID has been widely used in various fields since its emergence in the 1980s. The core of this model is to avoid introducing too many control variables that may lead to endogeneity and other problems. The study population is divided into a control group and an experimental group, then two dummy variables, policy trend and time trend, are introduced. After DID, the effect of the policy implementation is evaluated by observing whether one of the economic indicators of the experimental group changes significantly after the policy implementation.

The premise of the DID model is that the sample of the experimental group is selected to meet the statistical requirement of randomness. The release of national policies usually has a large impact and is generally implemented on a nationwide scale from point to point, so the policy launch event only meets the conditions of a quasi-experiment. Of course, the DID model can be infinitely close to random when selecting the experimental group samples, and the prior characteristics of the two groups are not required to be identical, so the DID model is gradually widely used in the field of “quasi-natural experiments”.

This study investigates the impact of new energy projects on economic development of cities based on this principle using a DID model. In this study, GDP, GDP per capita, employment rate, and industrial structure are selected to measure the economic development of cities. As a core indicator of a region’s economic aggregate, GDP reflects the economic scale and overall development level of a city. It is an important basis for assessing the level of economic development [31]. The GDP per capita indicator provides information on economic output per capita, which helps to measure the living standards and well-being of the population. GDP per capita can reveal whether economic growth is effectively translated into real benefits for residents, thus reflecting the inclusiveness of economic development [32]. Industrial structure reflects the diversity and modernization of economic activities. By analyzing the industrial structure of a city, we can understand the quality and sustainability of its economic development [33]. Employment rate is an important indicator of a city’s labor market condition, which directly affects residents’ income and social stability. High employment rates are usually associated with economic vitality [21].

Then, on this basis, the DID model is used to study the impact of the establishment of large-scale new energy projects on the economic development of the city. The formula is shown below.

$$Y^{psm}={\beta }_0+{\beta }_{1}time\times policy+{\beta }_i{X}_{it}+{\delta }_t+{\lambda }_i+{ϵ}_{it}$$

In this paper, $Y^{psm}$ is the explanatory variable of this study, which represents the economic development index of the city. In this paper, we will consider the effects of CDM projects on GDP, GDP per capita, employment rate, and industrial structure, respectively. $time$ is the point-in-time variable of CDM project completion, which is 0 before completion and 1 after completion. $policy$ is divided into two aspects, in which one aspect is the scale of cities’ CDM projects, and the other aspect is the number of cities’ CDM projects. $emission$ represents the carbon emission reductions of each city’s new energy CDM project, which is a continuous variable expressing the scale of the city’s CDM project, and the other $Num$ represents the number of city’s CDM projects. $X_{it}$ is a control variable, ${\delta }_t$ controls for time fixed effects, and ${\lambda }_i$ controls for individual fixed effects.

This study aims at a positive coefficient ${\beta }_1$, thus verifying the positive contribution of CDM projects to the economy of less developed coastal cities.

In

Table 2. Descriptive Statistics.

Variable	Obs.	Mean	SD.	Min	Median	Max
gdp	5704	149.120	253.609	3.040	73.067	3267.987
gdp_per	5599	3.523	4.333	0.076	2.243	53.662
labor	5611	0.121	0.116	0.003	0.085	1.473
indus	5606	51.067	15.349	0.840	52.295	94.820
popu	5618	4.340	3.746	0.145	3.641	110.984
road	5006	13.579	19.918	0.000	7.010	214.900
wage	5588	3.186	2.214	0.001	2.657	32.063
fdi	5053	731.226	1946.753	0.000	138.334	32,022.554
consume	5320	66.909	120.598	0.002	26.932	1341.390
fix	4997	78.937	121.556	0.740	37.567	1708.576
emission	5729	1.171	1.957	0.000	0.442	15.917
num	5729	9.519	13.111	0.000	5.000	88.000

, gdp is GDP of cities. gdp_per represents GDP per capita of cities. labor represents the number of cities’ employees. indus represents the industrial structure, expressed as the proportion of tertiary industry employees. popu represents the number of populations. road represents the paved road area. wage represents the average wage of employees. fdi represents the amount of actual foreign investment. consume represents retail sales of social consumer goods. fix is fixed asset investment. emission and num represent the scale of city’s CDM projects and the number of city’s CDM projects, respectively.

According to

Table 3. The impact of CDM projects on cities’ economic development.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
	ln_gdp	ln_gdp	ln_gdp	ln_gdp	lngdp_per	lngdp_per	lngdp_per	lngdp_per
post × emission	0.008 ***		0.008 ***		0.006 **		0.008 ***
	(5.61)		(5.67)		(2.25)		(5.67)
post × num		0.001 ***		0.001 ***		0.001 ***		0.001 ***
		(4.85)		(5.31)		(2.63)		(5.31)
t	−0.021	−0.022	−0.089 **	−0.091 **	−0.073	−0.075	−0.089 **	−0.091 **
	(−1.11)	(−1.14)	(−2.39)	(−2.45)	(−0.99)	(−1.01)	(−2.39)	(−2.45)
ln_popu			0.026 ***	0.027 ***			−0.974 ***	−0.973 ***
			(3.57)	(3.60)			(−131.50)	(−131.41)
ln_road			0.044 ***	0.044 ***			0.044 ***	0.044 ***
			(11.33)	(11.33)			(11.33)	(11.33)
ln_wage			0.018 ***	0.018 ***			0.018 ***	0.018 ***
			(2.92)	(2.95)			(2.92)	(2.95)
ln_fdi			−0.003 **	−0.003 **			−0.003 **	−0.003 **
			(−2.36)	(−2.32)			(−2.36)	(−2.32)
ln_consume			0.035 ***	0.036 ***			0.035 ***	0.036 ***
			(6.74)	(6.84)			(6.74)	(6.84)
ln_fix			0.077 ***	0.076 ***			0.077 ***	0.076 ***
			(18.42)	(18.35)			(18.42)	(18.35)
_cons	3.673 ***	3.686 ***	4.819 ***	4.858 ***	1.277	1.320	2.516 ***	2.555 ***
	(9.65)	(9.67)	(6.49)	(6.54)	(0.86)	(0.89)	(3.39)	(3.44)
cities’_fixed_effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
year_fixed_effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
N	5704	5704	4446	4446	5599	5599	4446	4446
r2	0.983	0.983	0.986	0.986	0.932	0.932	0.984	0.984
r2_a	0.98	0.98	0.98	0.98	0.93	0.93	0.98	0.98

Note: The values in parentheses are t-values, and ***, and **, represent significance levels of 1%, and 5%, respectively.

, the coefficient of city’s CDM project size is significantly positive at the 1% significance level, which implies that the expansion of city CDM project size can promote the enhancement of city GDP. Meanwhile, the increase in the number of city’s CDM projects also has a significant positive effect on the enhancement of city GDP, but the effect is weak. According to columns 7 and 8 of Table 3, both the expansion of the scale and the increase of the number of city’s CDM projects can drive the rise of GDP per capita. This is consistent with the findings of Hu et al. Their research indicates that CDM projects significantly enhance the development of regional GDP and per capita GDP, with longer implementation periods leading to greater positive impacts on the local economy [23].

As shown in Figure 4, there is no significant trend difference in the change in GDP before the policy shock.

According to

Table 4. The impact of CDM projects on city employment and industrial structure.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
	ln_labor	ln_labor	ln_labor	ln_labor	ln_indus	ln_indus	ln_indus	ln_indus
post × emission	−0.001		−0.000		0.017 ***		0.017 ***
	(−0.16)		(−0.04)		(5.35)		(4.34)
post × num		−0.002 **		−0.001 *		0.003 ***		0.003 ***
		(−2.42)		(−1.90)		(6.98)		(5.71)
t	0.169	0.173	0.160	0.162	−0.066	−0.073	−0.067	−0.073
	(1.34)	(1.37)	(1.42)	(1.44)	(−0.73)	(−0.81)	(−0.66)	(−0.72)
ln_popu			−0.894 ***	−0.894 ***			0.012	0.012
			(−39.74)	(−39.76)			(0.58)	(0.60)
ln_road			0.089 ***	0.090 ***			−0.055 ***	−0.056 ***
			(7.61)	(7.70)			(−5.21)	(−5.29)
ln_wage			−0.116 ***	−0.115 ***			0.116 ***	0.116 ***
			(−6.07)	(−6.06)			(6.81)	(6.83)
ln_fdi			−0.012 ***	−0.012 ***			0.007 **	0.007 **
			(−3.46)	(−3.39)			(2.33)	(2.31)
ln_consume			0.029 *	0.027 *			−0.047 ***	−0.045 ***
			(1.81)	(1.70)			(−3.32)	(−3.14)
ln_fix			−0.058 ***	−0.060 ***			0.014	0.015
			(−4.61)	(−4.74)			(1.24)	(1.30)
_cons	−5.601 **	−5.673 **	−4.408 *	−4.447 **	5.195 ***	5.335 ***	5.359 ***	5.470 ***
	(−2.22)	(−2.25)	(−1.96)	(−1.97)	(2.87)	(2.96)	(2.65)	(2.71)
city_fixed_effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
year_fixed_effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
N	5611	5611	4445	4445	5606	5606	4442	4442
r2	0.190	0.190	0.444	0.444	0.682	0.683	0.712	0.713
r2_a	0.14	0.14	0.40	0.40	0.66	0.66	0.69	0.69

Note: The values in parentheses are t-values, and ***, **, and * represent significance levels of 1%, 5%, and 10%, respectively.

, the expansion of CDM project scale and the increase in the number of CDM projects have a positive effect on industrial restructuring. However, the expansion of CDM project scale has no positive effect on employment, while the increase in the number of CDM projects has only a weak positive effect on employment.

Table 5. Impact of CDM projects on the coastal cities.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
	ln_gdp	ln_gdp	lngdp_per	lngdp_per	ln_labor	ln_labor	ln_indus	ln_indus
post × emission × coastal	0.014 **		0.014 **		0.094 ***		−0.056 ***
	(2.12)		(2.12)		(4.67)		(−3.11)
post × num × costal		0.001		0.001		0.006 **		−0.003
		(1.14)		(1.14)		(2.24)		(−1.48)
t	−0.088 **	−0.088 **	−0.088 **	−0.088 **	0.166	0.163	−0.070	−0.068
	(−2.37)	(−2.37)	(−2.37)	(−2.37)	(1.47)	(1.45)	(−0.69)	(−0.67)
ln_popu	0.027 ***	0.027 ***	−0.973 ***	−0.973 ***	−0.894 ***	−0.894 ***	0.012	0.012
	(3.59)	(3.59)	(−131.03)	(−130.98)	(−39.84)	(−39.77)	(0.60)	(0.60)
ln_road	0.045 ***	0.045 ***	0.045 ***	0.045 ***	0.089 ***	0.090 ***	−0.053 ***	−0.053 ***
	(11.55)	(11.55)	(11.55)	(11.55)	(7.64)	(7.64)	(−5.03)	(−5.04)
ln_wage	0.019 ***	0.019 ***	0.019 ***	0.019 ***	−0.114 ***	−0.115 ***	0.116 ***	0.117 ***
	(3.02)	(2.99)	(3.02)	(2.99)	(−5.98)	(−6.05)	(6.78)	(6.83)
ln_fdi	−0.002 **	−0.002 **	−0.002 **	−0.002 **	−0.012 ***	−0.012 ***	0.008 **	0.008 **
	(−2.08)	(−2.09)	(−2.08)	(−2.09)	(−3.37)	(−3.41)	(2.44)	(2.47)
ln_consume	0.034 ***	0.034 ***	0.034 ***	0.034 ***	0.027 *	0.028 *	−0.048 ***	−0.049 ***
	(6.49)	(6.50)	(6.49)	(6.50)	(1.72)	(1.76)	(−3.39)	(−3.42)
ln_fix	0.075 ***	0.075 ***	0.075 ***	0.075 ***	−0.055 ***	−0.057 ***	0.009	0.010
	(18.08)	(18.02)	(18.08)	(18.02)	(−4.37)	(−4.53)	(0.76)	(0.87)
_cons	4.806 ***	4.814 ***	2.504 ***	2.512 ***	−4.518 **	−4.459 **	5.432 ***	5.396 ***
	(6.46)	(6.46)	(3.36)	(3.37)	(−2.01)	(−1.98)	(2.69)	(2.67)
city_fixed_effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
year_fixed_effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
N	4446	4446	4446	4446	4445	4445	4442	4442
r2	0.985	0.985	0.984	0.984	0.447	0.444	0.711	0.710
r2_a	0.98	0.98	0.98	0.98	0.41	0.40	0.69	0.69

Note: The values in parentheses are t-values, and ***, **, and * represent significance levels of 1%, 5%, and 10%, respectively.

uses the difference-in-difference-in-difference (DDD) method and introduces the dummy variable of whether the city is a coastal city as a new interaction term to examine whether the impact of CDM project establishment on coastal cities is significantly different from that of inland cities. The results show that the expansion of CDM projects has a significant positive effect on the economic development of coastal cities and has a significant contribution to employment in coastal cities.

Table 6. Impact of CDM projects on cities at different levels of economic development.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
	ln_gdp	ln_gdp	lngdp_per	lngdp_per	ln_labor	ln_labor	ln_indus	ln_indus
post × emission × gdp	0.00002 **		0.00002 **		0.00010 ***		−0.00004
	(2.04)		(2.04)		(4.22)		(−1.64)
post × num × gdp		0.0000		0.0000		0.00001 **		−0.0000
		(1.18)		(1.18)		(2.27)		(−0.02)
t	−0.088 **	−0.088 **	−0.088 **	−0.088 **	0.164	0.162	−0.068	−0.067
	(−2.37)	(−2.38)	(−2.37)	(−2.38)	(1.46)	(1.44)	(−0.67)	(−0.66)
ln_popu	0.026 ***	0.027 ***	−0.974 ***	−0.973 ***	−0.895 ***	−0.894 ***	0.013	0.012
	(3.55)	(3.58)	(−131.04)	(−130.99)	(−39.89)	(−39.78)	(0.63)	(0.60)
ln_road	0.045 ***	0.045 ***	0.045 ***	0.045 ***	0.091 ***	0.090 ***	−0.053 ***	−0.053 ***
	(11.59)	(11.58)	(11.59)	(11.58)	(7.74)	(7.71)	(−5.06)	(−5.02)
ln_wage	0.019 ***	0.019 ***	0.019 ***	0.019 ***	−0.113 ***	−0.115 ***	0.116 ***	0.117 ***
	(3.04)	(3.00)	(3.04)	(3.00)	(−5.94)	(−6.02)	(6.79)	(6.84)
ln_fdi	−0.002 **	−0.002 **	−0.002 **	−0.002 **	−0.012 ***	−0.012 ***	0.008 **	0.008 **
	(−2.13)	(−2.13)	(−2.13)	(−2.13)	(−3.48)	(−3.47)	(2.51)	(2.50)
ln_consume	0.034 ***	0.034 ***	0.034 ***	0.034 ***	0.029 *	0.029 *	−0.049 ***	−0.049 ***
	(6.54)	(6.54)	(6.54)	(6.54)	(1.84)	(1.82)	(−3.46)	(−3.45)
ln_fix	0.076 ***	0.075 ***	0.076 ***	0.075 ***	−0.053 ***	−0.056 ***	0.009	0.010
	0.026 ***	0.027 ***	−0.974 ***	−0.973 ***	−0.895 ***	−0.894 ***	0.013	0.012
_cons	4.808 ***	4.815 ***	2.506 ***	2.513 ***	−4.499 **	−4.451 **	5.398 ***	5.367 ***
	(6.46)	(6.47)	(3.37)	(3.37)	(−2.00)	(−1.98)	(2.67)	(2.65)
city_fixed_effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
year_fixed_effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
N	4446	4446	4446	4446	4445	4445	4442	4442
r2	0.985	0.985	0.984	0.984	0.446	0.444	0.710	0.710
r2_a	0.98	0.98	0.98	0.98	0.40	0.40	0.69	0.69

Note: The values in parentheses are t-values, and ***, **, and * represent significance levels of 1%, 5%, and 10%, respectively.

also uses DDD method and introduces the city’s economic development level as a new interaction term to examine the impact of CDM projects on cities at different levels of economic development. The results show that the expansion of CDM projects has a significant positive effect on cities with high levels of economic development. And the development of CDM has a significant impact in promoting employment in cities with high levels of economic development.

4. Conclusions and Recommendations

4.1. Conclusions

Coastal cities have the natural resource endowment and location advantages to develop new energy. However, heterogeneity in the economic development of China’s coastal cities has led to differences in the outcomes of environmental regulatory policies and related programs. To elucidate the difference, this paper obtained 5074 CDM projects using crawler technology, conducted a causal identification through a quasi-experiment, and main conclusions are summarized as follows.

(1) Through the DID results, this paper finds that the development of CDM projects can promote the development of the city’s economy and lead to the upgrading of cities’ industries. On the one hand, CDM can help cities to obtain additional capital and advanced technology, which can promote the economic development of cities. On the other hand, CDM projects can promote the development of new energy projects, thus promoting industrial upgrading [8]. This finding highlights the positive impact of CDM projects on economic growth.

(2) The development of CDM projects can promote economic development and employment in coastal cities, which is more prominent in coastal cities with high levels of economic development. This is because CDM has promoted the development of new energy projects and new employment opportunities in Chinese cities through technology transfer and additional capital investment, thus driving the development of the local economy [34]. This result indicates that policymakers should prioritize economically robust coastal cities when promoting CDM projects to achieve more significant economic and social benefits, while also encouraging other regions to learn from successful experiences.

(3) This paper argues that the development of CDM projects have a mutually reinforcing effect with cities’ economic growth [35], and therefore CDM can better promote economic development and employment in less developed coastal areas when implemented in conjunction with economic development policies. This suggests that an integrated approach, combining environmental initiatives with economic strategies, may yield greater benefits for vulnerable regions, ultimately contributing to a more balanced and sustainable economic landscape.

4.2. Recommendations

Based on the main findings, policy recommendations are proposed as follows.

(1) The impact of CDM projects is different for each city, so the government should precisely put policies in place to set up projects where they are needed, otherwise it will lead to a waste of resources. The government should set up new energy project policies corresponding to the resource endowment of each place, so that the resource advantages of each place can be transformed into economic advantages.

(2) We found that CDM projects can promote the economic development of each region, but they can only promote employment in economically developed regions. Therefore, when the less developed coastal regions carry out CDM projects to promote their own economic development, they should also actively learn from the employment experience of developed regions to promote their own employment.

For example, Sichuan Province ranks first in the country in both the quantity and scale of household biogas systems in rural areas. Since 2010, Sichuan has actively developed greenhouse gas reduction projects related to rural biogas, planning to incorporate newly constructed household biogas digesters during the “Twelfth Five-Year Plan” period into its Clean Development Mechanism (CDM) projects. Between 2013 and 2015, Sichuan successfully facilitated carbon emissions trading in the international market, achieving a total traded carbon reduction volume of 1.82 million tons. Participants in the CDM projects received a total reduction benefit equivalent to 15.04 million RMB. As of September 2021, the number of developed carbon reduction pro-jects reached 530,000, with annual carbon reductions expected to exceed 1 million tons, significantly contributing to energy conservation and greenhouse gas reduction efforts in the province’s rural areas.