Next Article in Journal
Decoding Land Use Conflicts: Spatiotemporal Analysis and Constraint Diagnosis from the Perspectives of Production–Living–Ecological Functions
Previous Article in Journal
Mapping of Forest Species Using Sentinel-2A Images in the Alentejo and Algarve Regions, Portugal
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

The Impact of Low-Carbon City Construction on Urban Shrinkage: Evidence from China

School of Economics, Lanzhou University, Lanzhou 730000, China
*
Author to whom correspondence should be addressed.
Land 2024, 13(12), 2185; https://doi.org/10.3390/land13122185
Submission received: 2 November 2024 / Revised: 7 December 2024 / Accepted: 12 December 2024 / Published: 14 December 2024
(This article belongs to the Section Land Use, Impact Assessment and Sustainability)

Abstract

:
This paper uses Low-Carbon Pilot City (LCCP) as a quasi-natural experiment, 282 prefecture-level cities in China from 2007 to 2021, and models such as DID, SDM-DID, and DML to examine the impact of LCCP on urban shrinkage. Research shows that, first, LCCPs have effectively inhibited urban shrinkage, with pilot cities reducing urban shrinkage by 1.8% compared with non-pilot cities. Second, the LCCP may inhibit the city’s ability to shrink by reducing resource allocation efficiency, promoting technological innovation, and optimizing the living environment. Third, the urban shrinkage effect of the LCCP is heterogeneous depending on the economic region and whether the city is resource-based.
JEL Classification:
O18; Q54; R11

1. Introduction

The phenomenon known as “urban shrinkage” has surfaced since the establishment of the first cities. It soon spread worldwide and is currently a significant social and economic issue [1,2]. It is primarily characterized by population decline combined with unfavorable external development effects like wasteful use of land resources, empty housing, and economic downturns [3]. This significantly impacts and challenges the conventional urban and regional planning model based on the “growth scenario” and the global urban development cognition based on urbanization in the twenty-first century. Shrinking cities are rooted in globalization and localization, based on regional correlation, “competition and cooperation”, and background development constraints [4]. This results in the regional heterogeneity of shrinking cities in different development contexts, mechanisms, and response patterns. Urban shrinkage is not unique to developed countries. It is also happening in developing countries. According to UN-Habitat’s World Cities Report, 10 percent of the 1408 cities surveyed in developing countries lost population between 1990 and 2000. At the present stage in China, the regional development gap is becoming increasingly evident. There is differentiation inside and across cities; the flow of urban elements is mainly unrestricted, and the issue of urban shrinking is becoming more prevalent [5]. For the first time, “shrinking cities” were suggested in the National Development and Reform Commission’s “Key Tasks of New Urbanization Construction in 2019”, published in April 2019 [6]. Papers like “Key Tasks of New Urbanization Construction and Urban-Rural Integration Development in 2020” clarified the factual reality of diminishing cities as well as the reasonable expectations of planning response [7]. This indicates that China’s urbanization will now feature overall expansion and local contraction, with contraction emerging as a new problem for the country’s urban development.
In response to abrupt changes in climate, the United Nations developed the Sustainable Development Goals. Urban population, industry, and capital changes are directly linked to the development of climate change, as they represent a spatial carrier of the phenomenon. Urban shrinkage will inevitably result in slower industrial upgrading, less effective waste disposal, and a decrease in urban consumer demand, all of which are now strongly associated with sustainable development. Thus, “continuously optimizing the spatial layout and form of urbanization and accelerating the construction of new cities” is emphasized in the National Development and Reform Commission’s 2022 “Key Tasks for New Urbanization and Urban-Rural Integration Development [8]”. This development objective was put up in response to urbanization’s low degree of economic development, low levels of personal satisfaction, and climate change.
To achieve this, China’s National Development and Reform Commission launched a pilot initiative in 2010 called the “low-carbon city” [9,10]. The objective is to support low-carbon technology development and application, research and development, low-carbon building, management of China’s urban and rural areas, people’s lifestyles, and environmental quality. According to the low-carbon city evaluation standard system, a city can be identified as low-carbon if its low-carbon productivity index exceeds 20% of the national average. It also depends on the development of the city [11]. Consequently, can the Low-Carbon City Pilot Program (LCCP) reduce urban shrinkage? What are the internal transmission mechanisms? Does the LCCP have a correlated heterogeneous effect on urban shrinkage? Furthermore, as China places a high value on LCCP and has a sizable market for LCCP and related businesses, researching how LCCP affects urban shrinkage and its spillover impact can generate new solutions to the issue.
The following are this paper’s marginal contributions: First, the impact of LCCP on urban shrinkage has not been studied. At present, most studies focus on the overall form of urban shrinkage, and this study enriches the literature on low-carbon pilot policies and urban shrinkage. Second, by using the impact of exogenous policy on urban consolidation, the causal relationship between LCCP and urban shrinkage is effectively identified, which provides new evidence for the academic debate between LCCP and urban shrinkage and reveals the potential mechanism of LCCP’s influence on urban shrinkage. It will provide more targeted experience and a decision-making basis for the LCCP to formulate and implement relevant policies in China and other countries with similar development situations to China. Third, most of the existing literature ignores the existence of the spatial spillover effect, and ignoring the spatial correlation between cities will easily lead to the deviation of the model setting and overestimation of empirical results. If the assessment of the effects of the LCCP is biased, then the assessment of other indirect effects based on the transition to low-carbon cities will also be biased. While suppressing the spatial spillover of urban shrinkage, the SDID model satisfies the strict assumption of the individual processing stability of the DID model, solves the problem of overestimation of policy effects caused by ignoring spatial spillover, and improves the scientific validity and credibility of the empirical results.

2. Materials and Methods

2.1. Literature Review and Theoretical Analysis

Urban shrinkage is a complex issue that considers population, economy, society, and space. Therefore, developing a set of scientific evaluation techniques to quantify and analyze it is crucial [12]. From the standpoint of literature, most domestic and international researchers have enhanced it based on the findings of previous studies. For example, Zhang et al. [13] constructed a multi-dimensional urban contraction index system with nine indicators in three parts: population contraction, economic contraction, and social contraction. Since urban shrinkage involves the overall overview of production, living, and ecological levels at the city level, scholars have adopted various methods to measure urban shrinkage, such as principal component analysis, entropy value method, and entropy weight TOPSIS method. Nevertheless, only a few studies have examined how national pilot programs affect urban shrinkage when policies are implemented [14].
LCCP is a national policy exploration to cope with climate change, which has a mandatory intervention and restraint effect on ecological environment governance while considering economic and social benefits and helps restrain urban shrinkage. According to previous studies, the effective implementation of this policy will support China’s clean industry development, the optimization and modernization of the industrial structure, and the development of China’s ecological civilization and sustainable urban growth [15]. It will also reinforce the formulation of environmental protection policies.
There is relatively little literature on the relationship between low-carbon cities and urban shrinkage. Liu et al. [16] believe that, to pursue the objective of a low-carbon city, the LCCP needs to develop appropriate environmental management indicators as part of its environmental management plans. In the short term, if enterprises want to accomplish the purpose of an environmental protection policy, they must reallocate factor resources, update equipment, and increase investment in environmental protection. As a result, production factors will shift from environmental pollution management to productive activities. This will partially restrict the amount of capital, and labor businesses can effectively contribute to the production process and squeeze out productive investment. Simultaneously, the company’s unit labor output will decrease, impeding the advancement of economic quality and efficiency by affecting the factor productivity of companies. Labor resources will also be distributed to places with stricter environmental regulations, subject to the limitations of government pollution control. In contrast, capital and enterprises will be allocated to weaker areas. Therefore, the development of low-carbon cities in China may change the resource allocation structure, reduce resource allocation efficiency, and further aggravate the degree of urban shrinkage.
Different from the conclusion of this study, we believe that by innovating low-carbon technology, several industries simultaneously attempt to replace the original technology. The amount of funding allocated to clean and environmentally friendly production projects will increase, which will incentivize businesses to engage in autonomous innovation, accelerate the pace of green production technology research and development, foster the advancement of clean technology, lower energy and pollution levels, and enhance ecological capital [17]. At the same time, improving scientific and technological innovation capacity will also help curb urban shrinkage and other problems. Second, LCCP programs can enhance the capacity of pilot cities to oversee the environment [18]. They can also encourage high-energy-intensive businesses to invest in technological innovation and transform into green companies, which will advance China’s superior economicdevelopment [19]. Innovation in green technology involves low-carbon production and life, such as constructing green intelligent transportation systems. Finally, cities should focus on talent concentration, enhancing the innovation environment, and bolstering human capital while developing LCCP to boost low-carbon development [18,20]. The concentration and accumulation of human capital allows the city to innovate green technology and foster new knowledge and technology dissemination. However, it attracts new talent and social energy to the town, significantly preventing social and population decline [21].
By implementing production energy conservation, encouraging technical innovation, modifying the birth date structure, optimizing the energy consumption structure, and implementing green, energy-saving structures, the LCCP has enhanced urban living circumstances [22]. Many cities encourage citizens to adopt green and low-carbon consumption habits, publicize household energy consumption directly, take advantage of peer pressure and social norms as positive feedback loops, and improve their understanding of living a low-carbon lifestyle [23]. Simultaneously, numerous cities have actively created intelligent and green transportation networks [24].
The study presented above leads to the following two hypotheses being put forth in this paper:
Hypothesis 1.
LCCP will suppress both the level of urban shrinkage and the subdimension of urban shrinkage.
Hypothesis 2.
LCCP policies may curb urban shrinkage by reducing the efficiency of resource allocation, promoting technological innovation, and optimizing the living environment.
The impact mechanisms are shown in Figure 1.

2.2. Selection of Variables

2.2.1. Explained Variables

Urban shrinkage level (shrinklevel). As the body of knowledge about urban shrinkage grows, the consensus is that population, economy, and society are just a few factors that should be considered when measuring urban loss. Prior research has mainly focused on the extent of China’s urban shrinkage due to population structure adjustment, economic slump, and urban spatial decline [25]. They believe that one of the critical elements determining contraction is population decline [26], followed by economic recession and urban decline. Based on Zhang et al. [13]’s research, starting from population shrinkage, economic shrinkage, and social shrinkage, this paper comprehensively evaluates urban contraction with 15 fundamental indicators. Based on this, the data of fundamental indicators are dimensionalized using the range normalization approach, and the weights of indicators at each level are determined using the entropy method. Ultimately, these weights are used to fully estimate each city’s shrinkage level. Table 1 is a list of the particular indicators that were chosen.

2.2.2. Core Explanatory Variables

The LCCP implementation status of the City variable citylccpost indicates me in year t. Take one if City i has implemented the LCCP in year t; otherwise, take 0 [9]. For example, the third batch of the LCCP was determined in 2017; therefore, the value of 0 before 2017 is one after 2017, and the first and second batches of processing ideas are the same. Cities with severe data shortfalls are also excluded.

2.2.3. Control Variable

We chose seven major control factors that impact urban shrinkage to reduce the error caused by missing variables. Fiscal revenue and expenditure ratio: The ratio of public revenue to expenditure is expressed by fisc [27]; The degree of financial technology: techfina, a logarithmic calculation of the number of fintech companies in the region [28]; Economic openness: calculated as the region’s outward direct investment logarithm, denoted as lnfdi [29]; Research and development level: expressed as a percentage of GDP, defined as rd [30]; Financial development level: computed logarithmically from bank year-end deposit and loan balances to GDP, represented as lnfinance [5]; Advanced industrial structure: recorded as advanced when stated as the percentage of the tertiary industry’s added value to the GDP [31]; The level of education expenditure: using the logarithm of the number of people working in the city’s scientific and technical services sector, denoted as edu [32].

2.2.4. Instrumental Variable

This article uses inversion data from NASA satellite data to compare the average temperature of the first and second layers to determine the inversion situation. It uses the annual inversion days of each city, determined based on the first and second layers [33].

2.2.5. Mechanism Variable

  • Capital allocation efficiency. The green total factor productivity of prefecture-level cities is a measure of the effectiveness of capital allocation [34]. Hwang and Díez [35] calculated this productivity using the SBM-Malmquist-Luenberger index method.
  • Technological innovation. It is used in the Technology Innovation Index in the Fudan University 2017 China Urban and Industrial Innovation Report [9], which fully considers the value difference of patents at different stages. After adjusting for patent value, the stock index measures the regional intangible capital stock using the value of patents.
  • Ecological environment quality. Based on Xu et al. [36] from Beijing Normal University, a set of ecological environmental quality assessment models applicable to China’s regional scale was constructed, and the historical high-resolution ecological environmental quality data set of China was produced based on the model.

2.3. Methods

2.3.1. Methodology

Using the entropy weight method, the weights for each index are determined based on the entropy of each index. This model uses an objective weighting technique that can reduce the impact of human influence [37]. It measures a system’s disorder, which is commonly used in natural science to determine thermodynamic entropy. Information entropy is related to thermodynamics in practical applications, although it primarily focuses on quantifying the uncertainty of the system state [38]. The general view is that when the information entropy is large, the system’s structure will be balanced [39]. On the contrary, when the information entropy is small, it is manifested as structural imbalance, significant differences, or rapid changes [40]. This research measures urban shrinking using the entropy technique.

2.3.2. Model

The National Development and Reform Commission created 81 “low-carbon city” demonstration cities nationwide between 2010 and 2017 [9]. To determine how LCCP affects urban shrinkage in China, a high-dimensional fixed effect model is constructed using the DID model to study LCCP as a quasi-natural experiment [41]. The model is built as follows:
s h r i n k l e v e l i t = β 0 + β 1 c i t y l c c p o s t i t + β 2 c o n t r o l i t + μ i + γ t + ε i t
where c i t y l c c p o s t i t = t r e a t i × t i m e t .
In Formula (1), the subscripts i and t represent region and time, respectively. s h r i n k l e v e l i t is the dependent variable, which stands for the degree of urban contraction; c i t y l c c p o s t i t is the explanatory variable, which means the virtual variable of LCCP; and β 1 is the critical concern coefficient of this paper. If the coefficient is notably less than zero, it indicates that the LCCP can prevent urban decline from occurring. c o n t r o l i t symbolizes several chosen control variables, μ i demonstrates control over specific effects, γ t symbolizes control over time effects, and ε i t represents random disruption [11,21].

2.3.3. Data Source

This study’s data spans the years 2007 through 2021 and includes 282 cities at or above the prefecture level (certain cities with gravely inadequate data are excluded). Of these, 118 are pilot cities with low carbon emissions; the remaining 164 are not. The Fudan Institute of Industrial Development, School of Economics, Fudan University, and Kou and Liu [42] produced the FIND Report on City and Industrial Innovation in China, where the Urban Innovation Index originates. Quality of the ecological environment: The National Earth System Science Data Center and the National Science & Technology Infrastructure of China are the sources of the data, as well as the inversion data from NASA. Additional information utilized in this study is derived from the China City Statistical Yearbook. To maintain the continuity of the data when a few values are missing from the data for some cities, a linear interpolation approach is employed to replenish the missing values [9]. Table 2 displays the descriptive statistical findings for the variables in this investigation.

3. Results

3.1. Baseline Regression Results

The regression results in Table 3 demonstrate that the LCCP inhibits and negatively correlates with urban shrinkage. The estimated results of the LCCP are statistically negative at the 1% level when both control variables are absent and present. According to Column (2), the urban shrinkage level of pilot cities fell by 1.8% when several control variables were added, as opposed to non-pilot towns. This suggests that the LCCP had a mitigating effect on urban shrinkage. The result may be that the establishment of the LCCP provides a suitable living ecological environment for human beings, an excellent economic and innovative development environment for urban economic development, more policy facilities, and more efficient resource allocation for social development, thus encouraging cities to use LCCP-related policies to achieve the effect of restraining urban shrinkage. This needs to be further verified in the remainder of this article. Hypothesis 1 is proved. Under the dual pressure of improving the ecological environment and curbing urban shrinkage, implementing the LCCP can achieve the “win-win” goal of reducing greenhouse gas emissions and promoting urban development. The pilot policy will also reference other developing countries that deal with environmental and urban shrinkage issues.

3.2. DML Results

Concerning the work of Bia et al. [43], using the dual machine learning model (DML), construct the following model, re-regression, to compensate for the estimation error (Yang et al. [44]) caused by the curse of dimension, multiple collinearity, and the challenge of enumerating control variables in the empirical process of the DID model.
s h r i n k l e v e l i t = θ 0 c i t y l c c p o s t i t + g c o n t r o l i t + U i t
E U i t | c i t y l c c p o s t i t , c o n t r o l i t = 0
c i t y l c c p o s t i t = n c o n t r o l i t + W i t
E W i t | c o n t r o l i t = 0
This paper uses two algorithms, the neural network algorithm and the Lasso algorithm, to improve the persuasiveness of the conclusion on the impact of LCCP policy on urban shrinkage. This is performed to avoid the influence of model-setting bias on the conclusion (See Table 4). The DML model’s division ratio is set to 1:4. The fact that the findings resemble the baseline regression and show that the LCCP inhibits urban shrinkage while only slightly altering the amount of the policy effect is sufficient evidence for the validity of the initial conclusion.

3.3. Robustness Test Results

3.3.1. Parallel Trend Test and Dynamic Effect Identification

The underlying premise of the DID model, which is used to examine how LCCP affects s h r i n k   l e v e l , is that, before the introduction of LCCP, both the experimental and control groups must adhere to the common trend hypothesis. In other words, the trend of the urban shrinkage level in the pilot city should be consistent before the policy is implemented. Otherwise, it can be questioned whether the impact of LCCP causes the inhibition of urban shrinkage levels. Differences in their development trends cause these differences. This paper’s regression model (3) was set concerning Beck et al. [45].
s h r i n k l e v e l i t = β 0 + m 2 3 β m c i t y l c c p o s t i , t 0 + m + β 2 c o n t r o l i t + μ i + γ t + ε i t
The year t 0 in the formula above denotes the year city i established the LCCP city construction. The year m denotes the year m following the policy’s implementation, with values of −2, −1, 0, 1 2, 3, where negative values indicate before the implementation, and positive values indicate after the implementation; c i t y l c c p o s t i , t 0 + m symbolizes the dummy variable for the LCCP city construction “event”; β m is the main topic of this section, showing the variation in the degree of urban shrinkage in the year m of LCCP urban construction between the treatment group and the control group. When m < 0, if β m is not significantly different from 0, it suggests that before the policy was put into effect, there was no discernible difference in the development trend of the urban shrinkage level between the two groups of samples. The test of the equilibrium trend hypothesis is successful. It is not satisfied by the equilibrium trend hypothesis [46].
The results of this parallel trend test are shown in Figure 2 of the report. As a testament to the validity of the parallel trend hypothesis, the fact that there was no discernible difference in the rate of urban shrinkage between the experimental and control groups before the implementation of the pilot study supports the parallel trend hypothesis. Additionally, Figure 2 illustrates that with the introduction of the LCCP, urban shrinkage showed a gradually decreasing trend and a gradual weakening trend with the progress of the experiment. In general, the effect of this policy has been continuously strengthened. As a result, the influence of the LCCP on cities is continuous and does not exhibit a lag effect.

3.3.2. Placebo Test Results

As it is difficult for this paper to exhaust all control variables, there may be missing variables that may impact urban shrinkage [47]. Considering the non-randomness of policy impact and the heterogeneity among regions and avoiding the phenomenon of pseudo-regression [48], this essay refers to pertinent academic research, randomly sets the establishment time of the LCCP, sets new policy variables, and adds them to the benchmark model for regression. At the same time, the above operations were repeated 2000 times. If the regression coefficient of the corresponding 2000 times was insignificant, it indicated that the LCCP did not inhibit the urban shrinkage level after the random setting time. Two thousand placebo tests were performed, as shown in Figure 3. It is evident that the regression coefficient exhibits the properties of a normal distribution and is equally distributed on both sides of 0. This suggests that the inhibitory influence is not due to conventional random factors, and that the empirical results are robust [41].

3.3.3. Propensity Score Matching Test (PSM-DID) Results

The caliper nearest neighbor matching approach is employed in this work to minimize the disparity between the experimental and control groups and to rule out the potential impact of sample selection bias. On the one hand, panel data were treated as cross-section data for mixed matching; on the other hand, Böckerman and Ilmakunnas’s [49] method is used for phase-by-phase matching. The matched treatment and experimental groups were estimated using the DID technique (Equation (1)). The samples of prefecture-level cities with the most similar factors that may affect the selection of LCCP are matched, and DID regression is performed on the matched samples. The difference between the experimental and control groups was significantly reduced after matching. The benefits and drawbacks of matching are simultaneously displayed in the kernel density map of the two variables before and after matching. The effect of propensity score matching PSM is shown in Figure 4. Table 5 shows that the propensity scores of the control group and the policy experimental group become more comparable after matching, demonstrating the efficacy of matching. The findings in Column (1) of Table 5 indicate that the conclusions above remain valid even after accounting for PSM.

3.3.4. Synthetic DID Estimation

The ability to develop into an LCCP may be inherent to the systematic distinction between the experimental and control groups. It was precisely because of this imbalance that the city was chosen as the LCCP. Furthermore, the DID model is unable to account for the apparent regional variations in the duration and intensity of the LCCP, as well as potential regional variations in the demographic, economic, and social repercussions of the LCCP. This paper used the synthetic DID approach for robustness testing, as Arkhangelsky et al. [50] suggested, to investigate the results’ robustness. This approach is an estimation model that blends the synthetic control approach and the DID model. It can compute the treatment effect of each upgraded region and effectively address potential sample selection bias and policy endogeneity issues based on the synthetic DID approach. Table 6 presents the empirical results. The coefficient is reduced compared to the baseline regression results, but is still significant at the 1% level. This further demonstrates that before correcting the ex-ante trend, the article’s conclusion is still valid.

3.3.5. Other Robustness Tests

The entropy method measures the degree of population contraction, economic contraction, and social contraction, and the stability of the variables explained in the model (1) is tested. Table 7 (1), (2), and (3) show that even with the explanatory variables removed, the regression’s coefficient of determination remains significant, matching the outcome of a standard regression.

3.4. Endogeneity Test

To circumvent the issue of reverse causation, this article follows Chen et al. [51]’s methodology, uses MERRA-2 inversion data to compare the average temperature of the first and second layers, and computes the inversion data using NASA satellite data. Each city’s annual temperature inversion days, judged according to the first and second layers, are used as the instrumental variable. On the one hand, temperature inversion hinders the vertical flow of air, making it difficult for pollutants in the air to diffuse. Contaminants have a more challenging time breaking down when there is a temperature inversion, which impacts the effectiveness of the LCCP and satisfies the correlation of the instrumental variables. However, temperature inversion satisfies the exogenous criteria of instrumental variables because it is determined by geographical and meteorological factors rather than the random disturbance term. According to Table 8, the coefficient before the inversion variable in the first-stage regression is significantly negative, indicating a significant negative causal relationship between the temperature inversion days and the LCCP policy, which conforms to the correlation assumption of the instrumental variables. Similar to the baseline result, the two-stage estimation results demonstrate that the coefficient before the LCCP is strongly negative, showing that the LCCP policy significantly limits urban decline.

3.5. Mechanism Inspection

Both baseline regression and robust analyses show that the LCCP will inhibit urban shrinkage. On this basis, we will test the following three hypotheses proposed in the theoretical analysis part. Two models are constructed to verify the resource allocation effect and technological innovation effect:
G T F P i t = β 0 + β 1 c i t y l c c p o s t i t + β 2 c o n t r o l i t + μ i + γ t + ε i t
c r e a t i v e i t = β 0 + β 1 c i t y l c c p o s t i t + β 2 c o n t r o l i t + μ i + γ t + ε i t
First, Column (1) in Table 9 shows that the citylccpost regression result coefficient is −0.0038, which is negative and significant at the 10% level. This suggests that the LCCP will decrease the productivity of the green total factor, lessen the effectiveness of allocating urban resources, and subsequently alleviate the inhibitory effect on urban shrinkage. This could be the case because stringent environmental laws make it more likely for less productive businesses to close their doors and less likely for potentially polluting firms to open up shop. This will significantly reallocate resources within the industry and, to some extent, impede businesses’ ability to expand their production scale, thus impeding productivity growth. However, in the near run, tighter environmental regulations drive up the cost of production for businesses, eliminating their ability to use capital and labor effectively. This lowers green TFP and weakens the effect of the LCCP policy’s resource allocation on the inhibition of urban shrinkage.
Second, the LCCP can increase investment in urban technological innovation. Innovation is a powerful driving force for development, especially for a city. At the economic level, it plays a decisive role. This research presents an index of urban innovation ability, whose regression coefficient is 16.9977, significant at the 1% level, to examine its mechanism. This demonstrates how the LCCP may effectively encourage businesses’ technical advancement [52]. Simultaneously, promoting scientific and technological innovation can boost the percentage of innovative and renewable energy sources and clean production technologies in highly polluting industries [53]. LCCP stimulates the technological progress of enterprises by increasing investment in urban technological innovation, promoting clean energy production, and creating good living and working conditions for residents. Not only has the LCCP goal been achieved, but it has also suppressed population and social shrinkage, thus inhibiting urban shrinkage.
Third, the LCCP has carried out a relevant deployment of environmental regulations, aiming to provide favorable environmental support for pilot cities to achieve low-carbon economic development. Relevant environmental regulation policies can help cities experience issues regarding the quality of their natural environments during the urban shrinkage process. They can also lessen the effects of population and social shrinkage, provide residents with better living conditions, and eventually stop the shrinkage of cities. To determine whether applying the LCCP can prevent urban shrinkage by enhancing the quality of the ecological environment at the city level, this study examines how the LCCP improves ecological environment quality at the urban scale. This research creates a DDD model and incorporates ecological and environmental quality based on the model (7). The particular model is as follows:
s h r i n k l e v e l i t = β 0 + β 1 c i t y l c c p o s t i t × e c o l o g y i t + β 2 t r e a t t × e c o l o g y i t + β 3 t i m e i × e c o l o g y i t + β 4 c i t y l c c p o s t i t + β 5 c o n t r o l i t + μ i + γ t + ε i t
The regression results are shown in Table 10. This study reveals that the LCCP does not inhibit urban shrinkage by improving the ecological environment by having a coefficient of 0.1437, which is significant at the 1% level. Some efforts have been made to curb urban shrinkage despite government environmental controls. Because each pilot region’s low-carbon development planning is guided by green industries, projects, and production processes, China’s heavily polluting industries must develop slowly [51]. Due to this limitation, this paper cannot confirm whether environmental regulation policies can enhance the quality of the urban ecological environment and inhibit urban shrinkage based on its limited conclusions on optimizing the ecological environment’s quality. In the future, relevant data will be mined and tested to determine whether environmental regulation policies can help optimize urban ecological environment quality and inhibit urban shrinkage. Hypothesis 2 was tested.

3.6. Heterogeneity Analysis and Results Discussion

3.6.1. Economic Regional Heterogeneity

China’s regional development is highly unbalanced, especially in the eastern part of China, which has strong potential for economic growth due to its unique geographical advantages, institutional mechanisms, and other factors [54]. An analysis of the influence of regional heterogeneity of the LCCP on urban shrinkage is conducted in this study, in which the sample cities are divided into three geographical regions: eastern, western, and northeastern.
According to Table 11, Columns (1), (2), and (3) indicate that the LCCP alleviates urban shrinkage most effectively in the eastern region, followed by the northeast region, while its effect on the western region is significantly smaller than in the northeast and eastern regions. This clearly indicates that the level of local economic development will fully influence the LCCP. The eastern region has a higher level of informatization, science, technology, and economic growth than the western region, which will prevent urban shrinkage in the future.

3.6.2. Resource-Based Cities and Non-Resource-Based Cities

The previous analysis shows that the LCCP is well known for inhibiting urban search. Does the inhibitory effect of the LCCP still exist in different types of cities? A comparison of urban shrinkage inhibition effects between resource- and non-resource-based cities is presented in this paper [55] based on the National Sustainable Development Plan for Resource-Based Cities (2013–2020) and typical case studies.
Based on the heterogeneity analysis results, Table 12 shows a negative (−0.0006) but insignificant coefficient for citylccpost × resource, while a negative (−0.0214) coefficient for citylccpost × non-resource is significant at 1% and negative (−0.0214). As a result, non-resource-based cities can effectively avoid shrinking with the LCCP. It is because resource-based cities rely on resources too much, are under significant pressure from sustainable development, and lack the internal power to transform and grow, which prevents them from undergoing industrial transformation and upgrading, which is causing urban shrinkage to intensify [56]. As a result, the research on the inhibition effect of LCCP on resource-based cities has some limitations, and future studies and empirical analysis will be needed to prove whether LCCP alleviates environmental pollution effectively and curbs the shrinkage of resource-based cities effectively.

4. Discussion

When the LCCP positively inhibits local urban shrinkage, it also plays a role in surrounding urban shrinkage, the so-called spatial spillover effect [57].

4.1. Spatial Autocorrelation Test

When using the spatial measurement method, the spatial correlation of the main variables must first be tested. This paper uses global and local spatial autocorrelations and other methods to test the explained variables, and the results are displayed in Figure 5 and Figure 6. According to Figure 5, most cities are located in the first and third quadrants, indicating a spatial correlation between urban shrinkage and population growth. Using spatial econometrics to study urban shrinkage makes sense since the urban shrinkage level in the global Morans has a significant positive spatial correlation (Figure 6).

4.2. Construction of Spatial Metrology Model

Using the Moreland index, we can conclude that urban shrinkage is spatially positively correlated. This paper uses the spatial measurement method to determine whether the LCCP’s inhibitory effect on urban shrinkage is spatially spillover-related. Conversely, baseline regression can help alleviate endoplasmic problems to a great extent with the DID model. Despite this, macroscopic research overlooks the spatial correlation between geographical units and the differences between batches of LCCP [9]. Thus, this study improved the traditional DID by incorporating the SDID model based on Moser and Dube’s practice. To study the influence of LCCP on urban shrinkage, this paper adopted a multi-phase continuous SDM-DID model based on SDM since it is generally more general than the spatial Durbin model (SDM):
s h r i n k l e v e l i t = α 0 + ρ 1 W i j s h r i n k l e v e l i t + α 1 c i t y l c c p o s t i t + ρ 2 W i j c i t y l c c p o s t i t + α 2 c o n t r o l i t + ρ 3 W i j c o n t r o l j t + μ i + γ t + ε i t

4.3. Empirical Regression Results of Spatial Econometric Model

Based on the results in Table 13, it is evident that all the estimated coefficients of citylccpost are negative and pass the significance test of 1%. This shows that the LCCP significantly inhibits the city’s urban shrinkage level; that is, it inhibits the city’s urban shrinkage degree and validates Hypothesis 1. For the independent variable, W_citylccpost represents the spatial lag term. It is evident from columns (1) and (2) that the coefficient of the 01 matrix is positive but insignificant. On the other hand, the inverse distance space weight matrix coefficient is negative and passes the significance test by 1%. The results indicate that the degree of urban shrinkage is negatively affected by the LCCP. The spillover effect of the LCCP on shrinkage in neighboring cities will be felt when the policy is implemented. In other words, the LCCP affects the shrinkage in neighboring cities.
Using basic regression, the spatial effect is decomposed on this basis to assess the impact of the LCCP on the urban shrinkage of the city and the effect of the urban shrinkage degrees of the surrounding towns on the urban shrinkage degree. A spatial spillover effect is an indirect effect, that is, how LCCP implementation affects the urban shrinkage levels of neighboring cities. After implementing the LCCP in a particular region, the total effect is the average impact degree. In Table 14 and Table 15 (the 01 spatial weight matrix is used in Table 14, the inverse distance spatial weight matrix is used in Table 15, Column 1 represents direct effects, Column 2 represents indirect effects, and Column 3 represents total effects), the direct effects are shown in Column 1. Regarding the direct impact, the LCCP has a negative (−0.0149) effect on the city, which is significantly negative at the 1% level. It inhibits urban shrinkage in the town, proving that H1 is established. The indirect effects are negatively influenced by the LCCP coefficient, which passes the significance test at the 5% level and has a significant negative spillover effect on neighboring cities’ shrinkage levels. Regarding the overall effect, the LCCP coefficient (−0.0253) inhibits urban shrinkage, which is significant at the 1% level.

5. Conclusions

5.1. Conclusion and Policy Implications

This paper examines 282 Chinese prefecture-level cities from 2007 to 2021. An empirical analysis of the effects of urban shrinkage and LCCP uses multi-time point DID models, high-dimensional bidirectional fixed effects models, dual machine learning models, and spatial differential methods. The following conclusions are drawn:
First, the LCCP significantly inhibited urban shrinkage, which confirms Hypothesis 1.
Second, the LCCP curbs urban shrinkage by reducing the efficiency of resource allocation, promoting technological innovation, and optimizing the living environment, and Hypothesis 2 was tested.
Third, the impact of the LCCP on urban shrinkage varies from region to region.
Fourth, urban shrinkage levels across the country show an apparent spatial correlation with high and low agglomeration patterns.
Fifth, the negative spatial spillover effect of LCCP implementation means that the amount of urban shrinkage in a city is affected by the inhibitory effect of LCCP implementation in neighboring cities.
Based on the above conclusions, the following policy implications can be drawn:
First, actively promote the successful experience of the LCCP and gradually expand the pilot scope. At the same time, it is indispensable to establish a regular monitoring and appraisal mechanism to avoid imitation of the LCCPS, which leads to poor implementation.
Second, local governments should provide further financial and policy support for technology research and development, create a positive innovation environment, fully utilize the innovation compensation effect, and counteract the decline in total factor productivity caused by LCCP policies.
Third, regional economic and technical cooperation can be used to mitigate the negative spatial spillover effect of the LCCP on neighboring regions. Using inter-regional economic and technical cooperation, the policy spillover effect of the LCCP in the region can be further promoted, and its radiation effect on the surrounding areas can be exerted. To effectively facilitate regional economic and social development.

5.2. Discussions

The following two aspects can guide future work: First, as an extension, more attention can be paid to urban shrinkage and the impact of low-carbon pilots on it. Second, we found that the reason for the negative spatial spillover effect of low-carbon pilot implementation is that the amount of urban shrinkage in a city is affected by the inhibitory effect of low-carbon pilot implementation in neighboring cities (siphon effect). Therefore, researchers can continue to study how to reduce the negative spatial spillover effect of low-carbon pilot implementation. For example, whether low-carbon pilots also have negative spatial spillover effects on other urban forms (such as urban expansion and urban renewal).

Author Contributions

Methodology, M.H.; Software, M.H.; Investigation, M.H.; Data curation, Q.L.; Writing—original draft, B.L.; Writing—review & editing, B.L.; Project administration, B.L. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by Gansu Provincial Philosophy and Social Science Planning, grant number 2023YB006.

Data Availability Statement

Considering the non-public nature of the data, we do not have the authority to make the data public. However, if readers are particularly interested, we can share some unrestricted data; however, because the restricted and public data are combined, we will carefully consider when receiving readers’ requests for data.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Liu, Z.; Liu, S. Urban shrinkage in a developing context: Rethinking China’s present and future trends. Sust. Cities Soc. 2022, 80, 103779. [Google Scholar] [CrossRef]
  2. Lang, W.; Deng, J.; Li, X. Identification of “growth” and “shrinkage” pattern and planning strategies for shrinking cities based on a spatial perspective of the Pearl River Delta Region. J. Urban Plan. Dev. 2020, 146, 05020020. [Google Scholar] [CrossRef]
  3. Wang, S.; Liu, X.; Chen, X.; Song, M. An evaluative study of economic security from the perspective of land resource assets. Land Use Pol. 2024, 139, 107062. [Google Scholar] [CrossRef]
  4. Oswalt, P.; Rieniets, T. Atlas of Shrinking Cities; Hatje Cantz: New York, NY, USA, 2006. [Google Scholar]
  5. Zhang, X.; Zhang, Q.; Zhang, X.; Gu, R. Spatial-temporal evolution pattern of multidimensional urban shrinkage in China and its impact on urban form. Appl. Geogr. 2023, 159, 103062. [Google Scholar] [CrossRef]
  6. Wang, T.; Wu, K.; Yao, C.; Liu, X. Complex pathways to population shrinkage: Case study of Hegang City, China. Chin. Geogr. Sci. 2022, 32, 418–437. [Google Scholar] [CrossRef]
  7. Li, Z.; Chen, S.; Chang, X. Achieving clean energy via economic stability to qualify sustainable development goals in China. Econ. Anal. Policy 2024, 81, 1382–1394. [Google Scholar] [CrossRef]
  8. Cattaneo, A.; Adukia, A.; Brown, D.L.; Christiaensen, L.; Evans, D.K.; Haakenstad, A.; McMenomy, T.; Partridge, M.; Vaz, S.; Weiss, D.J. Economic and social development along the urban–rural continuum: New opportunities to inform policy. World Dev. 2022, 157, 105941. [Google Scholar] [CrossRef]
  9. Zeng, S.; Jin, G.; Tan, K.; Liu, X. Can low-carbon city construction reduce carbon intensity? Empirical evidence from low-carbon city pilot policy in China. J. Environ. Manag. 2023, 332, 117363. [Google Scholar] [CrossRef]
  10. Wu, Q.; Sun, Z.; Jiang, L.; Jiang, L. “Bottom-up” abatement on climate from the “top-down” design: Lessons learned from China’s low-carbon city pilot policy. Empir. Econ. 2024, 66, 1223–1257. [Google Scholar] [CrossRef]
  11. Wang, S.; Yang, D.Y. Policy Experimentation in China: The Political Economy of Policy Learning; National Bureau of Economic Research: Cambridge, MA, USA, 2021. [Google Scholar] [CrossRef]
  12. Mohsin, M.; Nurunnabi, M.; Zhang, J.; Sun, H.; Iqbal, N.; Iram, R.; Abbas, Q. The evaluation of efficiency and value addition of IFRS endorsement towards earnings timeliness disclosure. Int. J. Financ. Econ. 2021, 26, 1793–1807. [Google Scholar] [CrossRef]
  13. Zhang, Z.; Wang, J.; Feng, C.; Chen, X. Do pilot zones for green finance reform and innovation promote energy savings? Evidence from China. Energy Econ. 2023, 124, 106763. [Google Scholar] [CrossRef]
  14. Chen, W.; Wang, G.; Xu, N.; Ji, M.; Zeng, J. Promoting or inhibiting? New-type urbanization and urban carbon emissions efficiency in China. Cities 2023, 140, 104429. [Google Scholar] [CrossRef]
  15. Li, G.; Wen, H. The low-carbon effect of pursuing the honor of civilization? A quasi-experiment in Chinese cities. Econ. Anal. Policy 2023, 78, 343–357. [Google Scholar] [CrossRef]
  16. Liu, J.; Feng, H.; Wang, K. The low-carbon city pilot policy and urban land use efficiency: A policy assessment from China. Land 2022, 11, 604. [Google Scholar] [CrossRef]
  17. Li, J.; Liu, W.; Du, L.; Xiao, J. High-tech industry agglomeration and regional green development: An analysis of spatial Durbin model. Technol. Forecast. Soc. Chang. 2024, 123372. [Google Scholar] [CrossRef]
  18. Wen, S.; Liu, H. Research on energy conservation and carbon emission reduction effects and mechanism: Quasi-experimental evidence from China. Energy Policy 2022, 169, 113180. [Google Scholar] [CrossRef]
  19. Li, P.; Lu, Y.; Wang, J. Does flattening government improve economic performance? Evidence from China. J. Dev. Econ. 2016, 123, 18–37. [Google Scholar] [CrossRef]
  20. Song, M.; Zhao, X.; Shang, Y. The impact of low-carbon city construction on ecological efficiency: Empirical evidence from quasi-natural experiments. Resour. Conserv. Recycl. 2020, 157, 104777. [Google Scholar] [CrossRef]
  21. Liu, C.; Sun, W.; Li, P. Characteristics of spatiotemporal variations in coupling coordination between integrated carbon emission and sequestration index: A case study of the Yangtze River Delta, China. Ecol. Indic. 2022, 135, 108520. [Google Scholar] [CrossRef]
  22. Liu, Z.; Yu, C.; Qian, Q.K.; Huang, R.; You, K.; Visscher, H.; Zhang, G. Incentive initiatives on energy-efficient renovation of existing buildings towards carbon-neutral blueprints in China: Advancements, challenges and perspectives. Energy Build. 2023, 296, 113343. [Google Scholar] [CrossRef]
  23. Yang, F.; Shi, B.; Xu, M.; Feng, C. Can reducing carbon emissions improve economic performance–evidence from China. Economics 2019, 13, 20190047. [Google Scholar] [CrossRef]
  24. Chen, S.; Oliva, P.; Zhang, P. Air pollution and mental health: Evidence from China. AEA Pap. Proc. 2024, 114, 423–428. [Google Scholar] [CrossRef]
  25. Liu, Z.; Liu, S.; Song, Y. Understanding urban shrinkage in China: Developing a multi-dimensional conceptual model and conducting empirical examination from 2000 to 2010. Habitat Int. 2020, 104, 102256. [Google Scholar] [CrossRef]
  26. Shao, S.; Zhang, X.; Yang, L. Natural resource dependence and urban shrinkage: The role of human capital accumulation. Resour. Policy 2023, 81, 103325. [Google Scholar] [CrossRef]
  27. Afonso, A.; Jalles, J.T.; Venâncio, A. Fiscal decentralization and public sector efficiency: Do natural disasters matter? Econ. Model. 2024, 137, 106763. [Google Scholar] [CrossRef]
  28. Liu, X.; Zhao, S.; Ding, J.; Jia, Y. Role of Fintech adoption in the impact of sustainable policy intervention on enterprise transformation in resource-based cities: Evidence from China. Resour. Policy 2024, 88, 104443. [Google Scholar] [CrossRef]
  29. Truong, L.D.; Friday, H.S.; Pham, T.D. The Effects of Geopolitical Risk on Foreign Direct Investment in a Transition Economy: Evidence from Vietnam. J. Risk Financ. Manag. 2024, 17, 101. [Google Scholar] [CrossRef]
  30. Arza, V.; López, A.; Montes-Rojas, G.; Pascuini, P. In the name of TRIPS: The impact of IPR harmonisation on patent activity in Latin America. Res. Policy 2023, 52, 104759. [Google Scholar] [CrossRef]
  31. Xiong, M.; Zhang, F.; Zhang, H.; Wang, H. Digital economy, credit expansion, and modernization of industrial structure in China. Financ. Res. Lett. 2023, 58, 104500. [Google Scholar] [CrossRef]
  32. Yao, T.; Huang, Z.; Zhao, W. Are smart cities more ecologically efficient? Evidence from China. Sustain. Cities Soc. 2020, 60, 102008. [Google Scholar] [CrossRef]
  33. Jin, P.; Wang, S.; Yin, D.; Zhang, H. Environmental institutional supply that shapes a green economy: Evidence from Chinese cities. Technol. Forecast. Soc. Chang. 2023, 187, 122214. [Google Scholar] [CrossRef]
  34. Zhou, L.; Fan, J.; Hu, M.; Yu, X. Clean air policy and green total factor productivity: Evidence from Chinese prefecture-level cities. Energy Econ. 2024, 133, 107512. [Google Scholar] [CrossRef]
  35. Hwang, Y.K.; Díez, Á.S. Renewable energy transition and green growth nexus in Latin America. Renew. Sust. Energ. Rev. 2024, 198, 114431. [Google Scholar] [CrossRef]
  36. Xu, D.; Yang, F.; Yu, L.; Zhou, Y.; Li, H.; Ma, J.; Huang, J.; Wei, J.; Xu, Y.; Zhang, C.; et al. Quantization of the coupling mechanism between eco-environmental quality and urbanization from multisource remote sensing data. J. Clean Prod. 2021, 321, 128948. [Google Scholar] [CrossRef]
  37. Zhang, Y.; Khan, S.U.; Swallow, B.; Liu, W.; Zhao, M. Coupling coordination analysis of China’s water resources utilization efficiency and economic development level. J. Clean Prod. 2022, 373, 133874. [Google Scholar] [CrossRef]
  38. Camacho, M.; Romeu, A.; Ruiz-Marin, M. Symbolic transfer entropy test for causality in longitudinal data. Econ. Model. 2021, 94, 649–661. [Google Scholar] [CrossRef]
  39. Lynn, C.W.; Cornblath, E.J.; Papadopoulos, L.; Bertolero, M.A.; Bassett, D.S. Broken detailed balance and entropy production in the human brain. Proc. Natl. Acad. Sci. USA 2021, 118, e2109889118. [Google Scholar] [CrossRef] [PubMed]
  40. Chen, D.; Fang, C.; Liu, Z. Progress and major themes of research on urban shrinkage and its eco-environmental impacts. J. Geogr. Sci. 2023, 33, 1113–1138. [Google Scholar] [CrossRef]
  41. Chen, H.; Cheng, K.; Zhang, M. Does geographic proximity affect firms’ cross-regional development? Evidence from high-speed rail construction in China. Econ. Model. 2023, 126, 106402. [Google Scholar] [CrossRef]
  42. Kou, Z.; Liu, X. FIND Report on City and Industrial Innovation in China; Fudan Institute of Industrial Development, School of Economics, Fudan University: Shanghai, China, 2017; Available online: https://fddi.fudan.edu.cn/fddien/main.htm (accessed on 11 December 2024).
  43. Bia, M.; Huber, M.; Lafférs, L. Double machine learning for sample selection models. J. Bus. Econ. Stat. 2024, 42, 958–969. [Google Scholar] [CrossRef]
  44. Yang, J.C.; Chuang, H.C.; Kuan, C.M. Double machine learning with gradient boosting and its application to the Big N audit quality effect. J. Econom. 2020, 216, 268–283. [Google Scholar] [CrossRef]
  45. Beck, T.; Levine, R.; Levkov, A. Big bad banks? The winners and losers from bank deregulation in the United States. J. Financ. 2010, 65, 1637–1667. [Google Scholar] [CrossRef]
  46. Chen, C.; Lin, Y.; Lv, N.; Zhang, W.; Sun, Y. Can government low-carbon regulation stimulate urban green innovation? Quasi-experimental evidence from China’s low-carbon city pilot policy. Appl. Econ. 2022, 54, 6559–6579. [Google Scholar] [CrossRef]
  47. Wang, J.; Yang, Z.; Qian, X. Driving factors of urban shrinkage: Examining the role of local industrial diversity. Cities 2020, 99, 102646. [Google Scholar] [CrossRef]
  48. Su, X.; Tan, J. Regional energy transition path and the role of government support and resource endowment in China. Renew. Sust. Energ. Rev. 2023, 174, 113150. [Google Scholar] [CrossRef]
  49. Böckerman, P.; Ilmakunnas, P. Unemployment and self-assessed health: Evidence from panel data. Health Econ. 2009, 18, 161–179. [Google Scholar] [CrossRef]
  50. Arkhangelsky, D.; Athey, S.; Hirshberg, D.A.; Imbens, G.W.; Wager, S. Synthetic difference-in-differences. Am. Econ. Rev. 2021, 111, 4088–4118. [Google Scholar] [CrossRef]
  51. Chen, Y.; Chen, S.; Miao, J. Does smart city pilot improve urban green economic efficiency: Accelerator or inhibito. Environ. Impact Assess. Rev. 2024, 104, 107328. [Google Scholar] [CrossRef]
  52. Yang, S.; Jahanger, A.; Hossain, M.R. How effective has the low-carbon city pilot policy been as an environmental intervention in curbing pollution? Evidence from Chinese industrial enterprises. Energy Econ. 2023, 118, 106523. [Google Scholar] [CrossRef]
  53. Su, Y.; Fan, Q. Renewable energy technology innovation, industrial structure upgrading and green development from the perspective of China’s provinces. Technol. Forecast. Soc. Chang. 2022, 180, 121727. [Google Scholar] [CrossRef]
  54. Guo, Q.; Zhong, J. The effect of urban innovation performance of smart city construction policies: Evaluate by using a multiple period difference-in-differences model. Technol. Forecast. Soc. Chang. 2022, 184, 122003. [Google Scholar] [CrossRef]
  55. Yang, X.; Wang, W.; Wu, H.; Wang, J.; Ran, Q.; Ren, S. The impact of the new energy demonstration city policy on the green total factor productivity of resource-based cities: Empirical evidence from a quasi-natural experiment in China. J. Environ. Plan. Manag. 2022, 66, 293–326. [Google Scholar] [CrossRef]
  56. Pan, M.; Huang, L.; Song, D.; Qian, X. Can intellectual property rights protection reduce air pollution? A quasi-natural experiment from China. Struct. Chang. Econ. Dyn. 2023, 65, 210–222. [Google Scholar] [CrossRef]
  57. Ma, Z.; Zhou, G.; Zhang, J.; Liu, Y.; Zhang, P.; Li, C. Urban shrinkage in the regional multiscale context: Spatial divergence and interaction. Sust. Cities Soc. 2024, 100, 105020. [Google Scholar] [CrossRef]
Figure 1. The influence mechanism of LCCP on urban shrinkage.
Figure 1. The influence mechanism of LCCP on urban shrinkage.
Land 13 02185 g001
Figure 2. Dynamic effects.
Figure 2. Dynamic effects.
Land 13 02185 g002
Figure 3. Placebo test.
Figure 3. Placebo test.
Land 13 02185 g003
Figure 4. Effect of propensity score matching PSM.
Figure 4. Effect of propensity score matching PSM.
Land 13 02185 g004
Figure 5. Moran’s I scatter plot of local city shrinkage levels in 2007 and 2021.
Figure 5. Moran’s I scatter plot of local city shrinkage levels in 2007 and 2021.
Land 13 02185 g005
Figure 6. Global Moran Index of urban contraction levels from 2007 to 2021. Notes: (1) Standard errors are clustered at the city level and are indicated in parentheses; (2) *** represent statistical significance at the 1% levels.
Figure 6. Global Moran Index of urban contraction levels from 2007 to 2021. Notes: (1) Standard errors are clustered at the city level and are indicated in parentheses; (2) *** represent statistical significance at the 1% levels.
Land 13 02185 g006
Table 1. Three-dimensional evaluation system of the urban shrinkage index.
Table 1. Three-dimensional evaluation system of the urban shrinkage index.
DimensionalityIndexUnitIndex Attribute
Population shrinkageNatural growth rate%
Population density/
Birth populationThousands of people
Employed populationThousands of people
End-of-year total populationThousands of people
Economic shrinkagePrimary industry proportion%
Secondary industry proportion%
Tertiary industry proportion%
An average resident’s incomeTen thousand yuan
Amount of GDP per capitaTen thousand yuan
Fiscal revenue from general governmentTen thousand yuan
General public financial expenditureTen thousand yuan+
Social shrinkageAverage brightness of nighttime lights/
Unemployment rateTen thousand yuan
Unemployment rate%+
Table 2. Descriptive statistics of the variables.
Table 2. Descriptive statistics of the variables.
VariableObsMeanStd. Dev.MinMax
citylccpost42300.250.43301
shrinklevel42300.4360.0840.2280.699
fisc42300.4540.2230.0881.022
techfina42302.6571.71806.172
lnfdi42302.9461.63207.061
rd42308.2632.942011.719
lnfinance42305.3960.8753.7537.545
indstradvanced42300.9780.4930.2362.971
edu42303.623.1710.19817.157
Table 3. Basic results of the effect of LCCP on urban shrinkage.
Table 3. Basic results of the effect of LCCP on urban shrinkage.
Variables(1)(2)
citylccpost−0.0203 ***−0.0180 ***
(0.0044)(0.0039)
fisc −0.0118
(0.0135)
techfina −0.0065 ***
(0.0012)
lnfdi −0.0005
(0.0012)
rd −0.0022 ***
(0.0005)
lnfinance 0.0153 ***
(0.0044)
indstradvanced 0.0003
(0.0037)
edu −0.0037 ***
(0.0011)
Constant0.4413 ***0.4138 ***
(0.0011)(0.0214)
Year-FEYesYes
City-FEYesYes
Observations42304230
R-squared0.88270.8924
Notes: (1) Standard errors are clustered at the city level and are indicated in parentheses; (2) *** represent statistical significance at the 1% levels.
Table 4. DML results.
Table 4. DML results.
Variables(1)(2)
citylccpost−0.019 **−0.019 ***
(0.009)(0.004)
ControlsYesYes
Year-FEYesYes
City-FEYesYes
Observations42304230
Notes: (1) Standard errors are clustered at the city level and are indicated in parentheses; (2) *** and ** represent statistical significance at the 1%, and 5% levels, respectively.
Table 5. PSM-DID results.
Table 5. PSM-DID results.
Variables(1)
citylccpost−0.014 ***
(0.004)
Constant0.438 ***
(0.019)
ControlsYes
Year-FEYes
City-FEYes
Observations4111
R-squared0.911
Notes: (1) Standard errors are clustered at the city level and are indicated in parentheses; (2) *** represent statistical significance at the 1% levels.
Table 6. Robustness test of benchmark regression using the SDID method.
Table 6. Robustness test of benchmark regression using the SDID method.
Variables(1)(2)
citylccpost−0.0197 ***
(0.005)
−0.0171 ***
(0.005)
Controls NoYes
Year-FEYesYes
City-FEYesYes
Observations42304230
Se methodbootstrapbootstrap
Notes: (1) Standard errors are clustered at the city level and are indicated in parentheses; (2) *** represent statistical significance at the 1% levels.
Table 7. Test results for other robustness measures.
Table 7. Test results for other robustness measures.
(1)(2)(3)
VariablesPopulationEconomySocial
citylccpost−0.0109 *−0.0159 ***−0.0312 ***
(0.0058)(0.0039)(0.0089)
Constant0.7058 ***0.3107 ***0.4262 ***
(0.0271)(0.0272)(0.0432)
ControlsYesYesYes
Year-FEYesYesYes
City-FEYesYesYes
Observations423042304230
R-squared0.84920.92550.7846
Notes: (1) Standard errors are clustered at the city level and are indicated in parentheses; (2) *** and * represent statistical significance at the 1%, and 10% levels, respectively.
Table 8. Endogeneity test results.
Table 8. Endogeneity test results.
(1)(2)
VariablesCitylccpostShrinklevel
Intemp−0.0007 ***
(0.000)
citylccpost −0.2133 ***
(0.029)
Constant0.1183 **0.2852 ***
(0.059)(0.014)
ControlsYesYes
Year-FEYesYes
City-FEYesYes
Observations42304230
Notes: (1) Standard errors are clustered at the city level and are indicated in parentheses; (2) *** and ** represent statistical significance at the 1%, and 5% levels, respectively.
Table 9. Mechanism 1 and mechanism 2 regression results.
Table 9. Mechanism 1 and mechanism 2 regression results.
Variables(1)(2)
citylccpost−0.0038 *16.9977 ***
(0.0022)(5.3261)
Constant0.9867 ***104.7942 **
(0.0149)(40.7830)
Controls YesYes
Year-FEYesYes
City-FEYesYes
Observations42304230
R-squared0.09180.6490
Notes: (1) Standard errors are clustered at the city level and are indicated in parentheses; (2) *** ** and * represent statistical significance at the 1%, 5% and 10% levels, respectively.
Table 10. Mechanism 3 regression results.
Table 10. Mechanism 3 regression results.
Variables(1)
citylccpost × ecology0.1437 ***
(0.0289)
citylccpost−0.0434 ***
(0.0151)
Constant0.4222 ***
(0.0258)
ControlsYes
Year-FEYes
City-FEYes
Observations4230
R-squared0.8929
Notes: (1) Standard errors are clustered at the city level and are indicated in parentheses; (2) *** represent statistical significance at the 1% levels.
Table 11. Regional heterogeneity result 1.
Table 11. Regional heterogeneity result 1.
Variables(1)(2)(3)
citylccpost−0.0228 ***−0.0126 *−0.0225 **
(0.0054)(0.0071)(0.0100)
Constant0.3471 ***0.4357 ***0.5622 ***
(0.0343)(0.0373)(0.0302)
ControlsYesYesYes
Year-FEYesYesYes
City-FEYesYesYes
Observations2655930495
R-squared0.86250.91760.9395
Notes: (1) Standard errors are clustered at the city level and are indicated in parentheses; (2) *** ** and * represent statistical significance at the 1%, 5% and 10% levels, respectively.
Table 12. Results of heterogeneity analysis 2.
Table 12. Results of heterogeneity analysis 2.
Variables(1)(2)
citylccpost × resource−0.0006
(0.0056)
citylccpost × non-resource −0.0214 ***
(0.0047)
Constant0.4744 ***0.3569 ***
(0.0245)(0.0320)
ControlsYesYes
Year-FEYesYes
City-FEYesYes
Observations17102520
R-squared0.93700.8677
Notes: (1) Standard errors are clustered at the city level and are indicated in parentheses; (2) *** represent statistical significance at the 1% levels.
Table 13. SDM-DID results.
Table 13. SDM-DID results.
(1)(2)
VariablesW1W2
citylccpost−0.0142 ***−0.0182 ***
(0.0018)(0.0017)
W_citylccpost0.0005−0.0249 ***
(0.0027)(0.0093)
rho0.4695 ***−0.3047 ***
(0.0163)(0.0847)
sigma2_e0.0006 ***0.0008 ***
(0.0000)(0.0000)
ControlsYesYes
Observations42304230
R-squared0.27200.2112
Notes: (1) Standard errors are clustered at the city level and are indicated in parentheses; (2) *** represent statistical significance at the 1% levels.
Table 14. Decomposition results of spatial effects (W1).
Table 14. Decomposition results of spatial effects (W1).
Variables(1)(2)(3)
citylccpost−0.0149 ***−0.0104 **−0.0253 ***
(0.0018)(0.0041)(0.0044)
ControlsYesYesYes
Year-FEYesYesYes
City-FEYesYesYes
Observations423042304230
R-squared0.27200.27200.2720
Notes: (1) Standard errors are clustered at the city level and are indicated in parentheses; (2) *** and ** represent statistical significance at the 1% and 5% levels, respectively.
Table 15. Decomposition results of spatial effects (W2).
Table 15. Decomposition results of spatial effects (W2).
Variables(1)(2)(3)
citylccpost−0.0180 ***−0.0144 **−0.0325 ***
(0.0018)(0.0068)(0.0069)
ControlsYesYesYes
Year-FEYesYesYes
City-FEYesYesYes
Observations423042304230
R-squared0.21120.21120.2112
*** and ** represent statistical significance at the 1% and 5% levels, respectively.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Li, B.; Huang, M.; Li, Q. The Impact of Low-Carbon City Construction on Urban Shrinkage: Evidence from China. Land 2024, 13, 2185. https://doi.org/10.3390/land13122185

AMA Style

Li B, Huang M, Li Q. The Impact of Low-Carbon City Construction on Urban Shrinkage: Evidence from China. Land. 2024; 13(12):2185. https://doi.org/10.3390/land13122185

Chicago/Turabian Style

Li, Bowen, Meiying Huang, and Quan Li. 2024. "The Impact of Low-Carbon City Construction on Urban Shrinkage: Evidence from China" Land 13, no. 12: 2185. https://doi.org/10.3390/land13122185

APA Style

Li, B., Huang, M., & Li, Q. (2024). The Impact of Low-Carbon City Construction on Urban Shrinkage: Evidence from China. Land, 13(12), 2185. https://doi.org/10.3390/land13122185

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop