Next Article in Journal
Sustainable Work–Life Balance, Social Support, and Workload: Exploring the Potential Dual Role of Flexible Work in a Moderated Mediation Model
Previous Article in Journal
Sustainable Risk Mapping of High-Speed Rail Networks Through PS-InSAR and Geospatial Analysis
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Unlocking Synergies: How Digital Infrastructure Reshapes the Pollution-Carbon Reduction Nexus at the Chinese Prefecture-Level Cities

1
Newcastle Business School, The University of Newcastle, Newcastle 2300, Australia
2
School of Economics and Management, Jiangxi Normal University, Nanchang 330022, China
*
Author to whom correspondence should be addressed.
Sustainability 2025, 17(15), 7066; https://doi.org/10.3390/su17157066 (registering DOI)
Submission received: 29 June 2025 / Revised: 28 July 2025 / Accepted: 31 July 2025 / Published: 4 August 2025

Abstract

In the context of global climate governance and the green transition, digital infrastructure serves as a critical enabler of resource allocation in the digital economy, offering strategic value in tackling synergistic pollution and carbon reduction challenges. Using panel data from 280 prefecture-level cities, this study employs a multiperiod difference-in-differences (DID) approach, leveraging smart city pilot policies as a quasinatural experiment, to assess how digital infrastructure affects urban synergistic pollution-carbon mitigation (SPCM). The empirical results show that digital infrastructure increases the urban SPCM index by 1.5%, indicating statistically significant effects. Compared with energy and income effects, digital infrastructure can influence this synergistic effect through indirect channels such as the energy effect, economic agglomeration effect, and income effect, with the economic agglomeration effect accounting for a larger share of the total effect. Additionally, fixed-asset investment has a nonlinear moderating effect on this relationship, with diminishing marginal returns on emission reduction when investment exceeds a threshold. Heterogeneity tests reveal greater impacts in eastern, nonresource-based, and environmentally regulated cities. This study expands the theory of collaborative environmental governance from the perspective of new infrastructure, providing a theoretical foundation for establishing a long-term digital technology-driven mechanism for SPCM.

1. Introduction

The increasing impacts of climate change and pollution underscore the critical importance of emission reduction for sustainable development. With rapid industrialization and urbanization, China has become one of the world’s leading sources of pollutant and carbon emissions, facing immense pressure to curb emissions. As cities serve as key actors in environmental governance, identifying scientifically sound and effective strategies to advance SPCM at the urban level holds significant strategic importance for fostering green and sustainable socioeconomic development in China.
Against this backdrop, smart cities leverage policy guidance and resource integration to drive urban digital and intelligent transformation based on digital infrastructure development, offering new pathways and models for sustainable urban development. In recent years, smart city initiatives have proliferated globally. For example, in 2009, the United States embarked on smart city initiatives aimed at enhancing urban energy efficiency and environmental governance through digital technologies. Similarly, in 2005, the European Union implemented the i2010 strategy, seeking to improve urban operational efficiency by advancing communication technologies and building next-generation networks.
As the critical enabler of smart city development, digital infrastructure holds significant potential for enhancing urban operational performance, resource efficiency, and environmental governance capacity. However, key questions remain: How does digital infrastructure influence urban SPCM? What are the underlying mechanisms driving this relationship? Addressing these questions—by systematically examining whether digital infrastructure can achieve synergistic pollution-carbon reduction, clarifying its operational mechanisms, and mapping its impact pathways—is vital for advancing urban emission reduction and fostering sustainable urban development.
The literature has extensively examined SPCM and its contributing factors. Studies have confirmed that atmospheric pollutants and greenhouse gases share common sources and formation pathways [1], indicating that measures to reduce carbon emissions simultaneously decrease air pollution concentrations, thereby revealing inherent synergistic effects between pollution abatement and carbon mitigation. The literature primarily examines three key influencing dimensions: policy effects [2,3,4], technological effects, and structural effects [5,6,7]. Coordinated pollution-carbon policies, green technology innovation, and industrial and energy structure adjustments are key drivers of synergistic emission reductions.
Building upon existing research, this study identifies several critical gaps that warrant further exploration. While prior studies have examined the impact of digital infrastructure on SPCM from perspectives such as industrial development [8] and environmental regulation [9], few have specifically investigated this relationship under smart city policies or systematically elucidated the underlying mechanisms involved. Furthermore, although numerous studies have assessed the influence of digital infrastructure on urban SPCM, most lack rigorous causal identification and fail to incorporate multidimensional policy evaluations from heterogeneous perspectives. The current study addresses these limitations by employing robust endogeneity controls and comprehensive robustness tests, with a particular emphasis on deciphering the mechanistic pathways through which digital infrastructure enhances urban SPCM. These findings offer valuable empirical insights for advancing urban green and sustainable development.
Using panel data from 280 Chinese cities (2007–2022), this study examines the causal effect of digital infrastructure on urban SPCM. We measure SPCM performance via Super-SBM modeling and then apply multiperiod DID to assess the direct governance effects of digital infrastructure. Furthermore, mediation effect models are applied to uncover the operational pathways through which digital infrastructure influences urban SPCM. Our findings reveal a significant single-threshold effect of fixed-asset investment levels on this relationship, demonstrating that when investment surpasses a critical threshold, digital infrastructure’s governance effect on SPCM becomes attenuated. Finally, we conduct comprehensive heterogeneity analyses to explore how geographic location, resource endowments, and environmental regulations differentially shape the effectiveness of the digital infrastructure in achieving SPCM.
Our work makes three principal contributions to the extant research. First, adopting a novel infrastructure perspective, it thoroughly investigates the transmission mechanisms through which digital infrastructure facilitates urban SPCM, thereby enriching the theoretical framework of collaborative environmental governance. Second, through mechanistic examination, the study explores how digital infrastructure influences urban SPCM within the context of smart city pilot policies, significantly expanding upon previous research in this domain. Finally, building on urban characteristics, we examine the heterogeneous SPCM governance impacts of digital infrastructure across city types, providing multidimensional policy recommendations.
The paper proceeds as follows: Section 2 reviews the relevant literature, and Section 3 develops our theoretical framework, analyzing the impact mechanisms of digital infrastructure through direct effects and indirect channels. Section 4 details the econometric model and data. Section 5 presents the empirical results of the impacts of the digital infrastructure on the SPCM. Section 6 discusses findings in relation to existing research, and Section 7 concludes with policy implications.

2. Literature Review

2.1. Influencing Factors of SPCM

Building upon the co-origin hypothesis of carbon and pollution, greenhouse gases and atmospheric pollutants predominantly originate from extensive fossil fuel consumption, exhibiting similar emission characteristics that lend significant theoretical and practical value to collaborative research on pollution and carbon reduction. Scholars have primarily employed coupling coordination models and composite indicator approaches to measure SPCM performance. For example, Zhu et al. (2024) employed a coupling coordination model, which revealed improvements in the SPCM in the Yangtze River Delta, with an increase in high-synergy cities and a reduction in regional disparities [10]. Similarly, Abdullah and Usman (2022) adopted composite indicators to demonstrate significant synergistic mitigation potential between carbon and pollution reduction in the construction and transportation sectors [11]. The literature has focused predominantly on investigating mechanistic pathways and implementation strategies for achieving effective SPCM.
Current studies highlight three key pathways for pollution-carbon cogovernance: technological mitigation pathways, structural mitigation pathways, and policy coordination pathways [12]. With respect to technological and structural approaches, Dong et al. (2024) demonstrated that corporate green technology innovation significantly enhances resource utilization efficiency while simultaneously reducing both pollution and carbon emissions [13]. Liu et al. (2024) further established that shifting from traditional fossil fuels to cleaner energy sources through energy structure optimization effectively mitigates greenhouse gas and atmospheric pollutant emissions [14]. In the policy domain, initiatives such as low-carbon city pilot programs [15] and comprehensive big data experimental zones [16] have had positive impacts on coordinated pollution–carbon reduction. The academic consensus represented in the literature recognizes technological and structural mitigation pathways [17,18] as the two most critical approaches, with their synergistic effects having gained substantial empirical validation.

2.2. Environmental Dividends of Digital Infrastructure

The application of digital infrastructure has significant potential in facilitating both pollution and carbon emission reduction processes. The convergence of big data and AI technologies offers novel capabilities for tracking pollutants and controlling carbon emissions [19]. Digital platforms and analytical tools enable the effective integration and sharing of emission data [20], optimizing management systems and substantially improving urban environmental quality. With respect to pollution mitigation, Zou et al. (2023) employed a difference-in-differences (DID) approach leveraging China’s broadband strategy, revealing that network infrastructure development significantly reduces environmental pollution, with particularly pronounced effects on particulate matter reduction [21]. Complementing these findings, Li et al. (2025) employed spatial Durbin models, showing that digital infrastructure reduces local emissions while creating positive spillovers. contribute to environmental improvements in neighboring cities [22].
With respect to carbon mitigation, Song et al. (2024) analyzed China’s provincial panel data and demonstrated that digital infrastructure effectively reduces regional carbon emissions while exhibiting notable regional heterogeneity and spatial spillover effects [23]. Using urban panel data, Zhang et al. (2023) reported an inverted U-shaped relationship between digital infrastructure and carbon intensity [24]. Their findings suggest that carbon emission intensity may initially increase during the early stages of digital infrastructure deployment but subsequently significantly decrease as the infrastructure matures and becomes more sophisticated.

2.3. Influence of the Digital Infrastructure on the SPCM

The literature classifies the impact of digital infrastructure on emission control into three primary mechanisms. Studies have demonstrated that digital infrastructure optimizes industrial structures and enhances resource utilization efficiency, thereby simultaneously advancing both pollution abatement and carbon mitigation [25]. Zhang et al. (2025) verified that digital infrastructure significantly strengthens SPCM effects through its dual capacity to promote technological innovation and advance digital financial development [5]. Complementing these findings, Liang et al. (2025) revealed that digital infrastructure substantially reduces urban carbon emission intensity while generating measurable spatial spillover effects to surrounding regions [26].
The second category of studies presents alternative perspectives, with Mao et al. (2024) demonstrating that computing infrastructure development may increase energy consumption and associated carbon emissions, potentially undermining pollution and carbon reduction objectives [27]. Similarly, Nie et al. (2025) indicated that during initial construction phases, digital infrastructure expansion may generate additional pollution and carbon emissions due to technological limitations and regional disparities [28,29]. A third perspective reveals complex nonlinear relationships, where some scholars identify U-shaped effects of digital infrastructure on emission reduction, whereas others observe inverted U-shaped patterns in these environmental impacts.
While existing studies have extensively discussed the role of digital infrastructure in pollution reduction and carbon mitigation separately, significant gaps remain in understanding its mechanisms for achieving synergistic governance of these dual objectives. This research context makes investigating how digital infrastructure facilitates urban SPCM particularly valuable, as it not only enriches the theoretical framework of collaborative environmental regulation but also provides practical insights for achieving sustainable development goals.

3. Policy Background and Research Hypotheses

3.1. Policy Background

The rapid advancement of global urbanization has intensified urban challenges, including traffic congestion, environmental degradation, and resource scarcity, positioning smart city development as a pivotal strategy to address these multidimensional issues. By integrating IoT, big data analytics, and AI technologies, smart city programs enhance intelligent urban management through precision governance approaches. China’s national smart city program was formally institutionalized in 2012 when the MOHURD issued interim administrative measures for national smart city pilot projects, initiating the first batch of 90 pilot cities at the prefectural and county levels. This was subsequently expanded through additional batches implemented in 2013 and 2014. Current official statistics indicate that the Ministry has approved 290 cities as designated smart city pilots to date. These systematically implemented pilot projects are projected to exert substantial and enduring impacts across multiple dimensions of China’s socioeconomic development trajectory. The spatial distribution of the smart city pilot zones is presented in Figure 1.

3.2. Research Hypothesis

3.2.1. Analysis of Direct Effects

The development of digital infrastructure significantly contributes to enhancing urban SPCM, which is manifested primarily across three dimensions: digital production, digital governance, and digital lifestyles. In terms of digital production, the incorporation of cloud computing, AI, and other digital applications—supported by digital infrastructure—into enterprise production and management has driven the transition toward digitalized production management systems [30,31]. This shift helps reduce excessive resource consumption and pollutant emissions during production and management processes. By enabling enterprise digital transformation and fostering industrial digitization, digital infrastructure improves resource utilization efficiency and reduces pollution emissions [32], directly contributing to urban green development.
From the perspective of digital governance, the adoption of digital infrastructure, such as information technologies, digital platforms, and intelligent tools, significantly enhances information flow [21,33], facilitating efficient communication and collaboration among governments, businesses, the public, and other stakeholders. This strengthens interconnections and interactions between governance entities, leading to optimized resource allocation and improved urban environmental governance efficacy. With respect to digital lifestyles, digital infrastructure provides the public with convenient tools and platforms. The proliferation of internet and social media platforms has made dissemination and access to environmental information easier [34], enabling the public to obtain such information more efficiently. This accessibility helps raise environmental awareness among citizens, thereby advancing urban emission reduction efforts.
Hypothesis 1.
Digital infrastructure can help promote urban SPCM.

3.2.2. Analysis of Energy Effects

Digital infrastructure enhances energy utilization efficiency through the integration, transfer, and application of resources via information technologies and communication networks—a critical pathway for achieving urban synergistic pollution-carbon mitigation. On the production front, through real-time monitoring of industrial energy consumption [35], digitalized energy systems allow enterprises to accurately detect inefficiencies and improve energy utilization patterns [36,37]. This operational refinement directly reduces both pollutant discharges and carbon emissions. Simultaneously, smart sensor networks, IoT systems, and big data analytics platforms generate high-precision energy consumption and environmental monitoring data [38]. Such technological capacities allow for scientifically informed designs of urban transport and energy infrastructure, improving energy efficiency while furthering emission reduction objectives.
Hypothesis 2.
Digital infrastructure facilitates urban SPCM through enhanced energy utilization efficiency.

3.2.3. Analysis of Economic Agglomeration Effects

Digital infrastructure dismantles information flow barriers and significantly enhances information circulation efficiency, enabling interfirm resource integration and collaborative innovation [39]. This capability attracts enterprise agglomeration, fostering industrial clusters that generate economic agglomeration effects. Such spatial concentration facilitates both production scale economies and centralized pollution control, improving environmental governance efficiency through optimized resource utilization. Furthermore, economic agglomeration promotes shared access to innovative equipment, advanced technologies, and managerial expertise among colocated firms [40], effectively reducing innovation costs and stimulating collaborative innovation. This sharing mechanism activates technology diffusion and knowledge spillover effects [41], increasing regional innovation capacity and ultimately advancing synergistic pollution-carbon mitigation governance.
Hypothesis 3.
Digital infrastructure promotes urban SPCM through economic agglomeration effects.

3.2.4. Analysis of Income Effects

Digital infrastructure development significantly enhances average labor productivity and fosters the growth of value-added industries [42], whereas telecommuting and online platforms expand employment opportunities [43], collectively contributing to urban economic development and increased household incomes. The increase in per capita disposable income further facilitates urban SPCM through two complementary pathways: consumption structure transformation and corporate product innovation. With respect to consumption patterns, higher disposable incomes drive a shift from subsistence-oriented food expenditures to development-focused service consumption, encompassing education, healthcare, and cultural activities [44,45]. This structural transition reduces material resource intensity and correspondingly decreases emissions.
From the innovation perspective, elevated incomes strengthen green consumption preferences, with higher income demographics demonstrating greater willingness to pay premium prices for environmentally friendly products and services [46]. This market dynamic incentivizes firms to pursue technological innovation, develop sustainable offerings, and minimize their environmental footprint, thereby significantly improving urban environmental quality.
Hypothesis 4.
Digital infrastructure promotes urban SPCM through income effects.

3.2.5. Analysis of Threshold Effects

The capital-intensive nature of digital infrastructure, which encompasses broadband deployment and data center construction, makes fixed-asset investment a critical metric of development scale and intensity, which in turn affects its pollution–carbon mitigation performance. In the initial development phase, increased investment levels extend the scale and spatial reach of digital infrastructure, thereby strengthening its positive environmental impacts [37,47]. However, as investment continues to grow and digital infrastructure approaches saturation, its marginal contribution to pollution and carbon reduction gradually diminishes [48,49]. When investment exceeds a certain threshold, the environmental benefits derived from digital infrastructure begin to weaken. Consequently, beyond optimal investment levels, additional funding may no longer effectively enhance the synergistic pollution–carbon mitigation effects.
Hypothesis 5.
The enhancement effect of the digital infrastructure on the urban SPCM exhibits threshold characteristics contingent on fixed-asset investment levels.
The conceptual framework of the study is visualized in Figure 2. To address climate change and environmental degradation, our research systematically examines the direct effects, indirect transmission mechanisms, and threshold characteristics of digital infrastructure on urban SPCM. This investigation establishes both a theoretical framework and actionable pathways for building green and sustainable urban development systems.

4. Research Design

4.1. Sample Selection and Data Sources

This study analyzes data from 280 Chinese prefecture-level cities. Considering the availability of data, the observation interval is from 2007 to 2022. The smart city pilot list was compiled from three batches of official announcements by China’s MOHURD. This method is used to reduce the carbon reduction collaborative efficiency of carbon dioxide emission data from the global emissions EDGAR database for atmospheric research. Other data were obtained from China city statistical yearbooks, China environmental yearbooks, China environmental statistical bulletins, and China city construction statistical yearbooks.

4.2. Model Design

4.2.1. Benchmark Model

Given that, years may show different trends. It is necessary to include year fixed effects to control the characterizing factors that change with the year but not with the individual. Similarly, since each city has its own characteristics and its pollution emission and carbon emission levels are different, it is necessary to include city fixed effects to control for potential characteristics that change with individual cities, forming two-dimensional fixed effects.
S P C M i t = α 0 + β 1 D I D i t + β 2 X i t + u i + v t + ε i t
In the above formula, the SPCM represents synergistic pollution and carbon mitigation. β 1 captures the policy treatment effect as the core explanatory variable’s coefficient. If it is significantly positive, digital infrastructure can significantly promote urban SPCM. X refers to a series of control variables, u is the city fixed effect, v is the year fixed effect, and ε is the random disturbance term. Unless otherwise specified, the same applies to other models.

4.2.2. Threshold Effect Model

Adopting the fixed-asset investment level (fix) as a threshold variable, this study constructs a single-threshold regression model:
S P C M i t = δ 1 p o s t i t I p o s t i t Y 1 + δ 2 p o s t i t I p o s t i t > Y 1 + μ X i t + u i + v t + ε i t
where Y 1 is the threshold value. I · is the indicator function, and the value is 1 if the parenthesis condition is true; otherwise, the value is 0 .

4.3. Definition of Key Variables

4.3.1. SPCM

This study adopts the Super-SBM model, which eliminates radial and directional biases while incorporating slack variables, to measure the SPCM. The evaluation index system for the SPCM is constructed across three dimensions: input indicators, including urban energy consumption, year-end employment, and fixed capital stock; desirable outputs, represented by real GDP; and undesirable outputs, comprising concentrations for pollution efficiency and emissions for carbon efficiency.
The fixed capital stock is calculated via the perpetual inventory method with the formula K t = I t P t + 1 δ t K t 1 , where K t denotes the fixed capital stock in period t (with the base-year stock calculated as 2006 fixed-asset investment divided by 10%), I t represents fixed-asset investment in period t , P t stands for the fixed asset price index in period t (proxied by the urban consumer price index), and δ t is the depreciation rate set at 9.6%.
Urban energy consumption is estimated by first calculating the correlation coefficient between provincial-level nighttime light data and energy consumption and then applying this coefficient to city-level nighttime light data to derive city-level energy consumption. The emission data are obtained from EDGAR, with the original point data converted to raster format via R software 4.4.3 and subsequently aggregated by region. Under the constant returns to scale (CRS) assumption, the Super-SBM model is specified as follows:
ρ = m i n 1 m i = 1 m s i x i k 1 q 1 + q 2 r = 1 q 1 s r + y r k + k = 1 q 2 s t y y t k
s . t . j = 1 , j k n x i j λ j s i x i k
j = 1 , j k n y r j λ j s r + y r k
j = 1 , j k n y t j λ j s t y y t k
1 1 q 1 + q 2 r = 1 q 1 s r + y r k + k = 1 q 2 s t y y t k > 0
λ , s , s + > 0 , j = 1 , j k n λ j = 1
In Equation (3), ρ for the SPCM, n represents the number of decision-making units, m and q 1 , q 2 represent the number of elements in the input, expected output and unexpected output, respectively, x i k , y r k ,     y t k represents the vectors of the input, expected output and unexpected output, respectively, and s i , s r + ,     s t y represents the corresponding relaxation variables. The input factors and the desired output and nondesired output indicators selected for this study are summarized in Table 1.
Based on the measured SPCM, this study further illustrates the spatiotemporal distribution of this efficiency in Figure 3. In 2010, before smart city policy implementation, the overall synergistic performance in terms of pollution and carbon reduction across pilot regions remained relatively low. By 2014, following the policy’s enactment, the governance levels for both pollutant emissions and carbon emissions in these regions improved. The data for 2018 and 2022 demonstrate that several years after policy implementation, the coordinated governance capacity in the pilot areas significantly strengthened. Overall, from 2010 to 2022, cities exhibited a positive trend in the SPCM, with both carbon mitigation and pollution control achieving notable progress.

4.3.2. Digital Infrastructure

The core explanatory variable (DID) is a dummy variable of the smart city pilot policy, consisting of a group dummy variable (Treat) and an annual dummy variable (Post). This variable is assigned a value of 1 only when the city is on the pilot list after the policy implementation; otherwise, it is assigned a value of 0 . The smart city pilot policy has been in place for 14 years since its implementation, providing sufficient time coverage and data granularity to use this policy as a proxy variable for digital infrastructure.

4.3.3. Control Variables

A range of variables that may influence the SPCM are controlled for in the empirical model [10,29,39,50]. Population density (pop) is measured by the resident population per urban area. Educational attainment (edu) is represented by the ratio of education expenditure to general government fiscal expenditure. Environmental regulation (env) is measured by the comprehensive utilization rate of general solid waste. Mobile phone penetration (mobile) is measured as the average number of mobile phones per capita. The financial development level (fin) is indicated by the ratio of year-end financial institution deposit balances to regional GDP. The informatization level (inf) is captured through the ratio of per capita telecommunications business volume to per capita GDP. The employment structure (emp) is measured as the proportion of tertiary industry employment in total employment. Fixed-asset investment intensity (fix) is represented by the ratio of fixed-asset investment to regional GDP. Industrial structure upgrading (ind) is calculated as the weighted average of the proportion of value added by the three industries to regional GDP. Healthcare capacity (hosp) is measured by the number of hospital beds per 100 inhabitants. Technological capacity (sci) is quantified as the logarithm of total patent applications.

4.3.4. Mechanism Variables

This study investigates the mechanistic pathways through which digital infrastructure influences urban SPCM via three distinct channels: the energy effect (energy), the economic agglomeration effect (eagg), and the income effect (income) [42,51,52]. The energy effect is operationalized using GDP per unit of energy consumption. Economic agglomeration effects are measured by GDP per capita, whereas income effects are quantified through disposable income per capita.

4.3.5. Threshold Variable

This study employs the fixed-asset investment level (fix) as the threshold variable. Digital infrastructure development requires substantial capital investment, where fixed-asset investment intensity reflects a region’s financial commitment to digital infrastructure construction and serves as a proxy for its development maturity. Table 2 provides the formal definitions of all the primary variables.
The descriptive statistics for all the study variables are reported in Table 3. We employed linear interpolation for limited missing values and excluded cases with substantial missing data to maintain result reliability. The cleaned dataset exhibits no significant data gaps, with all the variables demonstrating numerically and economically plausible values.

5. Empirical Test and Analysis

5.1. The Impact of the Digital Infrastructure on the SPCM

Table 4 presents the estimated results of the synergistic effects of digital infrastructure on urban pollution reduction and carbon emission reduction. As shown in Table 4, column (1) lists the results with control variables included and without the fixed effects of cities and years. The coefficient of DID is 0.023, which is significant at the 1% level. After adding the urban fixed effect in column (2), the coefficient still remains significantly positive. In column (3), after individual fixed effects and annual fixed effects are simultaneously incorporated, digital infrastructure has a statistically significant positive effect on the synergistic effect of urban pollution reduction and carbon emission reduction. In an economic sense, the results in column (3) show that digital infrastructure can significantly increase a city’s SPCM by 1.5%, and the empirical results support Hypothesis 1.
On the basis of data from 2007 to 2022, this paper analyzes the data of 280 prefecture-level cities in China by constructing a DID model and concludes that digital infrastructure can promote the synergistic effect of pollution reduction and carbon reduction in cities, which is consistent with the research results of Zhang et al. (2025) and Li et al. (2025) [5,22].
Digital infrastructure development offers strategic potential for urban pollution–carbon coreduction in the current digital transformation era. Digital infrastructure enables the precise monitoring and control of resource and energy utilization in both industrial production and government administration [37], facilitating optimal resource allocation and enhancing energy efficiency, which effectively reduces pollutant and carbon emissions. Furthermore, the agglomeration effects and knowledge spillovers generated by digital infrastructure foster interfirm collaborative innovation [53,54], increasing regional innovation capacity to advance green technology development and providing critical support for next-generation pollution–carbon reduction technologies.

5.2. Validity Test of the DID Model

5.2.1. Parallel Trend

The parallel trend assumption is a prerequisite for using a multiperiod DID model, which requires that the experimental group and the control group have the same or similar changing trends before being affected by policy shocks. The results in Figure 4 show that the policy estimation coefficient is not significant before the implementation of the smart city pilot policy but becomes significantly positive after the policy is implemented. This indicates that there was no significant difference between the experimental group and the control group before the implementation of the smart city pilot policy, meeting the parallel trend hypothesis test. These findings suggest that the construction of digital infrastructure has a significant positive effect on the synergistic effect of pollution reduction and carbon emission reduction.

5.2.2. Placebo Test

To examine whether the impact of digital infrastructure on the synergy of pollution reduction and carbon emission reduction is caused by undetectable random factors, 60 cities were randomly selected as the intervention group, false pilot variables were artificially constructed, and on the basis of the settings of Model (1), after 500 regressions, the p value distribution and kernel density of the estimated values of the regression coefficients were plotted. The results are shown in Figure 5. The randomly assigned estimated values are concentrated around 0, and the p values of most estimated coefficients are greater than 0.1. The mean of the false estimated coefficients is close to 0, which is much smaller than the true estimated value of 0.015 in the benchmark regression. This finding indicates that the impact of digital infrastructure on the synergy of pollution reduction and carbon emission reduction is genuine and effective, verifying the robustness of the regression results.

5.3. Robustness Analysis

5.3.1. PSM-DID

Since the smart city pilot policy does not constitute a strict natural experiment, self-selection bias may arise in evaluating its policy effects. To address this issue, we employ the PSM-DID method for robustness checks. Specifically, we construct a logit model using control variables as covariates and perform propensity score matching through three approaches: nearest-neighbor matching, caliper matching, and kernel matching. The matched samples are then analyzed via the baseline regression model. As shown in Table 5, the coefficients of digital infrastructure remain significantly positive across all specifications, with no statistically significant differences from the baseline regression results. This confirms the robust effect of digital infrastructure in promoting urban SPCM.

5.3.2. Excluding the Impact of Other Policies

During the study period, two potentially confounding policy interventions—the Broadband China Strategy (BCS) pilot launched in 2013 and the National Big Data Comprehensive Pilot Zone (NBDCPZ) initiative implemented in 2016—may influence urban synergistic pollution-carbon mitigation outcomes. To isolate these effects, we incorporate both policy dummies into our baseline regression specification. As Table 6 shows, the digital infrastructure coefficient remains statistically significant and positive after controlling for these concurrent policies, providing robust evidence that digital infrastructure development independently enhances the urban SPCM.

5.3.3. Mitigating the Impact of Nonrandom Selection

The multiperiod DID approach ideally requires random assignment of treatment and control cities. However, the designation of pilot zones is inherently nonrandom, which may compromise the accuracy of the estimation results. To address this concern, our baseline regression incorporates interaction terms between city characteristics and time trends, effectively controlling for the influence of inherent urban attributes on synergistic pollution–carbon mitigation effects. As Table 6 shows, the coefficient for digital infrastructure remains statistically significant at the 1% level, with a positive sign. This finding confirms the validity of our multiperiod DID approach even after mitigating nonrandom selection concerns, thereby supporting the robustness of our core results.

5.3.4. Reducing the Impact of Extreme Values

To further mitigate potential bias from outliers in the baseline regression, we implement a 1% winsorization on both the dependent variable and all continuous control variables before re-estimating Model (1). As presented in Table 6, the digital infrastructure coefficient remains statistically significant at the 5% level with consistent directional effects compared with the baseline estimates, confirming the robustness of our primary findings.

5.3.5. Dual Machine Learning Model

In the digital economy era, nonlinear relationships between variables are more common, but traditional linear regression assumes linear relationships between variables, and the model setting is biased, making the estimates less robust. Dual machine learning, on the other hand, can effectively eliminate bias and ensure the unbiasedness of the estimated coefficients by constructing a model using residuals. This paper constructs the following dual machine learning model on the basis of the robustness of dual machine learning model validation:
S P C M i t = ψ o D I D i t + g ( X i t ) + U i t
E [ U i t | X i t | D I D i t ] = 0
D I D i t = f ( X i t ) + V i t
E [ V i t | X i t ] = 0
where ψ 0 is the estimation coefficient of interest in this paper, and the remaining variables are consistent with the reference regression. The difference is that the learning algorithm estimation form of the control variable set is g X i t , and Equation (7) indicates that the conditional mean of the random error term is 0 . Table 7 shows that the estimated coefficients of the data infrastructure variable are significantly positive, which indicates that the results are robust.

5.3.6. Bacon Decomposition

In multiperiod DID designs, variation in treatment timing across samples may introduce bias in two-way fixed-effects (TWFE) estimators, potentially yielding incorrect treatment effect estimates or even reversed causal interpretations. On this basis, this paper adopts the robust estimators proposed by Borusyak as robust estimators. For multistage DID, the control group consisted of three types: those who had never been treated (Never), those who had been treated before (Late), and those who had been treated before (Early). If early is used as the control group, the traditional estimator may be biased. When the weight of the estimated coefficient of this category is large and contrary to the actual effect of the policy, it will have a very large effect on the result. In this paper, Bacon decomposition is used to test the bias of the estimator, and the results are shown in Table 8. Among them, the late-T vs. early-C category indicates late-T as the treatment group and early-T as the control group. The results show that the coefficient of this category is negative, contrary to the policy effect, but the weight proportion is relatively small (5.5%), indicating that there is no serious bias in the heterogeneous treatment effect and that the results are relatively robust.

5.3.7. Multiphase and Multi-Individual DIDM Estimation Subsection

In traditional bidirectional fixed-effect estimation, the heterogeneity of therapeutic effects in the group (class) and time dimensions may lead to biased results. Therefore, this paper adopts a multiperiod and multi-individual DIDM model for robustness testing. The results are shown in Table 9. The results indicate that the average treatment effect of policy conversion is 0.016, which is significantly positive at the 1% level.
We further estimate the dynamic treatment effects of the smart city pilot policy over five pretreatment periods and eight posttreatment periods, with the event study visualization presented in Figure 6. The results demonstrate statistically insignificant policy effects during the pretreatment phase, followed by gradually emerging and statistically significant impacts after policy implementation. These dynamic effect patterns remain fully consistent with our baseline regression estimates, thereby providing additional empirical validation for the robustness of our primary findings.

5.3.8. Stacked DID Estimation

Given the potential heterogeneity in treatment effects across different pilot cohorts and implementation years, baseline estimates may suffer from bias. To assess the robustness of the policy effects, we construct a stacked dataset by sequentially matching each treatment cohort (cities implementing smart city policies) with control groups drawn exclusively from never-treated cities. This stacked estimation approach mitigates biases inherent in conventional two-way fixed-effects specifications. The results are shown in Table 9. The coefficient of digital infrastructure is significantly positive at the 5% level, indicating that the benchmark regression results are robust.

5.4. Endogenous Test

To address potential endogeneity arising from unobserved omitted variables and nonrandom selection into smart city pilots, we implement an IV approach using river density as our instrument for digital infrastructure development. The IV estimation results are presented in Table 10. Column (1) reports the first-stage estimates, showing a statistically significant coefficient of 2.527 for the instrument at the 1% level. Column (2) presents the second-stage results, with the digital infrastructure coefficient remaining statistically significant (0.035, p < 0.01). The Kleibergen-Paap F statistic of 18.815 exceeds conventional weak instrument thresholds, confirming the instrument’s validity. These results demonstrate that digital infrastructure has a statistically robust positive effect on urban synergistic pollution–carbon mitigation even after endogeneity concerns are accounted for.

5.5. Mechanism Analysis

The regression results in Table 11 show that digital infrastructure can achieve urban SPCM by improving energy utilization efficiency, promoting economic agglomeration and increasing people’s disposable income, and the estimated coefficients are all significant at the 1% level.
The essence of synergistic pollution-carbon mitigation governance lies in optimizing resource and energy utilization through multistakeholder collaboration to reduce pollution sources and carbon emissions simultaneously. Central to this approach is enhancing energy efficiency as a fundamental pollution prevention strategy at the source. Digital infrastructure applications, including IoT systems, big data analytics platforms, and energy management systems, empower both enterprises and governments to transform traditional production and governance models into digitized operational frameworks [55]. This technological transition reduces excessive resource consumption and pollutant emissions throughout production and administrative processes while significantly improving energy efficiency, thereby facilitating the realization of urban synergistic pollution-carbon mitigation. These findings provide empirical validation for Hypothesis 2.
Economic agglomeration yields scale effects that enable infrastructure, technology and resource sharing among firms [56], lowering innovation costs while increasing innovation capacity. The resulting knowledge spillovers facilitate access to innovation resources [57,58], accelerating green technology development and adoption for combined pollution–carbon reduction. Agglomeration further enhances environmental governance efficiency through centralized pollution treatment. These findings provide empirical validation for Hypothesis 3.
An increase in per capita disposable income facilitates urban pollution–carbon reduction through both demand- and supply-side mechanisms. Higher incomes drive consumption patterns toward service-oriented expenditures, stimulating growth in the tertiary sector, which demonstrates characteristically low pollution and energy intensity [59], thereby reducing urban energy consumption and emissions. Furthermore, affluent demographics exhibit stronger environmental consciousness and greater demand for green products [60], creating market pressures that incentivize firms to develop eco-friendly innovations, ultimately lowering both pollutant discharges and carbon emissions. These findings provide empirical validation for Hypothesis 4.
Table 12 presents the effect decomposition results. The mechanism analysis identifies three significant transmission channels: energy effects (contributing 19.02% of total effects), economic agglomeration (eagg) effects (22.11%), and income effects (20.31%). These findings demonstrate that economic agglomeration effects make the most substantial contribution, highlighting their critical role in facilitating the digital infrastructure’s promotion of SPCM. However, both energy effects and income effects remain statistically and practically significant, representing important complementary pathways through which digital infrastructure achieves urban SPCM.

5.6. Threshold Effect Analysis

To examine whether fixed-asset investment levels exhibit threshold effects on synergistic pollution–carbon mitigation, we conduct threshold effect tests with 300 bootstrap replications, with the results presented in Table 13. The analysis reveals a statistically significant single-threshold characteristic for digital infrastructure development at the 95% confidence level. Figure 7 displays the corresponding likelihood ratio function plot, where the estimated threshold value falls within the confidence interval, confirming the validity of the identified threshold.
Table 14 presents the threshold regression results. When fixed-asset investment levels are below the threshold value of 0.285, digital infrastructure has a statistically significant positive coefficient of 0.135 (p < 0.01) for the SPCM. However, when investment exceeds 0.285, the coefficient diminishes to 0.011. These results demonstrate a distinct single-threshold characteristic in digital infrastructure promotion by the SPCM, with fixed-asset investment intensity serving as the threshold variable, thereby confirming Hypothesis 5.
This threshold pattern may stem from differential marginal returns at various investment stages. When fixed-asset investment remains below the threshold (0.285), additional funding expands the scale and coverage of digital infrastructure, thereby increasing its environmental benefits [48,49]. However, as investment surpasses this critical level, infrastructure approaches saturation, and cities achieve optimal resource utilization, causing the marginal contribution of digital infrastructure to SPCM to gradually diminish [29,50]. Consequently, the pollution–carbon reduction effects weaken when investment exceeds the identified threshold.

5.7. Further Discussion: Heterogeneity Analysis

5.7.1. Geographic Regions

Given the substantial regional disparities in economic development, resource endowments, and openness across China, the effect of digital infrastructure on urban SPCM may exhibit substantial heterogeneity. Table 15 presents location-based heterogeneity tests, revealing stark contrasts between the eastern and central-western regions. For eastern cities, the estimated coefficient of 0.042 (significant at the 1% level) reflects their successful smart city initiatives that enhance the capacity of digital infrastructure to provide pollution governance and environmental protection services, thereby improving SPCM efficiency. In contrast, central-western regions show statistically insignificant negative coefficients, likely due to resource constraints, weaker economic dynamism, and limited innovation capacity, which collectively hinder digital infrastructure deployment and consequently impede SPCM achievement.

5.7.2. Resource Endowment

The inherent differences in implementation difficulty between environmental governance and carbon reduction inevitably influence how smart city digital infrastructure affects pollution and emission mitigation. Following China’s national sustainable development plan for resource-based cities (2013–2020), we conduct a comparative analysis between resource-based and nonresource cities to assess how digital infrastructure differentially influences pollution-carbon synergy effects. Table 16 reveals distinct patterns: digital infrastructure development significantly enhances urban SPCM in nonresource cities but has no statistically meaningful impact on resource-based cities.
This divergence likely stems from resource cities’ ecological challenges, where excessive resource extraction exceeds their environmental carrying capacity, exacerbating ecological damage, pollution, and resource waste, all of which compound governance difficulties. Conversely, nonresource cities demonstrate stronger agglomeration capacities for talent and capital, enabling them to leverage advanced technologies for sustainable development and regional green growth.

5.7.3. Environmental Regulation

Given that environmental protection efforts may differentially affect pollution-carbon synergy across cities, we classify urban areas into environmental priority cities and nonpriority cities on the basis of China’s 11th five-year National Environmental Protection Plan issued by the State Council in 2007. As shown in Table 17, environmentally prioritized cities demonstrate markedly stronger SPCM outcomes than their nonpriority counterparts do. This disparity likely stems from three interrelated factors in priority cities: stricter environmental regulations and energy-saving subsidies implemented by local governments, greater resident environmental awareness leading to active participation in conservation initiatives, and a deeper public understanding of pollution–carbon reduction imperatives. Conversely, nonpriority cities face structural constraints, including incomplete environmental legislation and greater reliance on energy-intensive, polluting industries, which currently limits the potential of digital infrastructure to enhance urban SPCM governance.
To address deteriorating air pollution, China implemented two control zone policies starting in 1998. As demonstrated in Table 18, digital infrastructure significantly enhances SPCM within these TCZ cities. This amplified impact stems from the policy’s dual mechanisms: first, it compels local enterprises to reduce emissions, driving out low-technology, high-pollution manufacturers through industrial relocation; second, it creates strong incentives for green technology innovation, accelerating corporate environmental upgrading. These combined forces make the contribution of digital infrastructure to urban SPCM particularly pronounced in TCZ regions.

6. Discussion

This study demonstrates that digital infrastructure can effectively promote urban SPCM, which aligns with the findings of Zhang et al. (2025) and Li et al. (2025) [5,22]. The literature has established empirical evidence supporting the constructive role of digital infrastructure in enhancing environmental governance outcomes. For example, Zou et al. (2023), using China’s broadband strategy as a quasinatural experiment, demonstrated that the development of network infrastructure significantly reduces environmental pollution, particularly particulate matter emissions [21]. Similarly, Li et al. (2025) provide evidence that digital infrastructure reduces both local air pollution and generates beneficial spillover effects across neighboring regions [22]. These conclusions provide robust support for our findings.
However, some studies present contrasting perspectives. Mao et al. (2024) reported that the construction of computing infrastructure may lead to increased energy consumption, thereby increasing carbon emissions and potentially hindering pollution and carbon reduction goals [27]. Additionally, Nie et al. (2025) [28] highlighted that the initial phase of digital infrastructure expansion, due to technological limitations and regional disparities [30], could result in elevated pollution and carbon emissions. Therefore, the strategic rollout of digital infrastructure must carefully consider local technological capabilities and regional development conditions to maximize its environmental benefits.
Second, digital infrastructure influences urban SPCM through three primary pathways: energy efficiency improvement, economic agglomeration effects, and income effects. Notably, the economic agglomeration effect contributes most significantly to the overall impact, highlighting the crucial role of economies of scale and collaborative innovation in driving sustainable development. Zhang et al. (2021) demonstrated that industrial concentration facilitates energy sharing and large-scale application of clean technologies, thereby accelerating low-carbon industrial transformation and promoting carbon emission reduction [41]. These findings align with the literature on the role of digital infrastructure in fostering economic-environmental synergies.
Third, this study identifies a single-threshold effect of fixed-asset investment levels in the relationship between digital infrastructure and the SPCM. When investment exceeds a certain threshold, the marginal benefits for pollution and carbon reduction diminish. This suggests that while substantial investment in digital infrastructure is necessary to promote urban SPCM, beyond this point, additional investment yields progressively smaller environmental returns. This conclusion is corroborated by related research. Wang et al. (2025) examined the threshold effect of digital infrastructure saturation on urban eco-efficiency and reported that when digital infrastructure investment intensity surpasses a critical threshold, its elasticity coefficient for improving urban eco-efficiency decreases [48]. The identified threshold effects reveal that optimizing environmental returns from digital infrastructure requires carefully calibrated scaling of investments and strategic phasing of deployment.
Fourth, the influence of digital infrastructure on urban SPCM displays marked regional variations, with stronger effects in eastern regions, nonresource-based cities, key environmental protection cities, and cities within the two control zones. These differential effects stem from variations in economic development levels, resource endowments, and policy implementation effectiveness. Eastern regions, with their advanced economic and technological foundations, possess greater capacity to harness digital infrastructure for environmental benefits. In contrast, the central and western regions face greater challenges due to resource constraints and weaker innovation capabilities. Similarly, nonresource-based cities, which are less dependent on traditional polluting industries, are better positioned to benefit from digital infrastructure in achieving SPCM. These results highlight the importance of developing differentiated policy interventions that consider the distinct socioeconomic and environmental contexts of various regions and city categories.

7. Conclusions and Policy Implications

7.1. Research Conclusions

Achieving SPCM constitutes a pivotal pathway for sustainable urban development. By utilizing panel data from 280 Chinese prefecture-level cities (2007–2022), this study employs the Super-SBM model to quantify SPCM performance, followed by panel regression, mediation effect, and threshold effect analyses to comprehensively examine the impacts of digital infrastructure, transmission mechanisms, and nonlinear characteristics of the SPCM. The principal findings are as follows: (1) The baseline regression results demonstrate a statistically significant positive coefficient for digital infrastructure at the 1% level, confirming its robust promoting effect on urban SPCM—a conclusion that withstands multiple robustness checks and endogeneity treatments. (2) Energy efficiency gains, economic agglomeration effects, and sustainable income growth emerge as the three primary mechanisms through which digital infrastructure enhances SPCM. (3) Digital infrastructure’s impact exhibits a single-threshold effect relative to fixed-asset investment levels, with diminishing marginal returns beyond the threshold. (4) Heterogeneity analysis reveals more pronounced effects in eastern regions, nonresource-based cities, key environmental protection cities, and two control zones.

7.2. Theoretical Contributions

The present research makes three substantial contributions to the literature. First, it adopts a novel infrastructure perspective to thoroughly investigate the transmission mechanisms through which digital infrastructure facilitates urban SPCM, thereby enriching the theoretical framework of collaborative environmental governance. Second, while previous research has extensively examined the impacts of digital infrastructure on pollution reduction and carbon emission separately, there has been a notable lack of systematic investigations into its synergistic effects on both objectives simultaneously. Furthermore, few studies have explored the mechanisms by which digital infrastructure influences urban SPCM against the backdrop of smart city pilot policies. By analyzing both the direct and indirect effects of digital infrastructure on urban SPCM under these pilot policies, this research substantially expands upon existing knowledge. Third, through comprehensive heterogeneity analysis on the basis of urban characteristics, this study examines variations in the impact of digital infrastructure on SPCM across different city types and discusses the implications of these differences from multiple perspectives. These findings yield practical policy recommendations aligned with the sustainable development needs of various urban settings, providing policymakers with actionable insights to maximize the environmental benefits of digital infrastructure deployment.

7.3. Practical Value

SPCM serves as a fundamental driver of high-quality development and sustainable urban transformation, where digital infrastructure deployment offers a mutually reinforcing solution for both environmental preservation and economic progress. Building upon our empirical findings, we propose three key policy recommendations for global implementation.
National governments, particularly in developing economies, must first formally acknowledge the strategic importance of digital infrastructure in advancing urban SPCM and enabling green economic transitions. This recognition should translate into concrete actions, including the formulation of comprehensive digital infrastructure development plans, the optimization of institutional frameworks to support infrastructure rollout, and the facilitation of multistakeholder collaboration across geographical, administrative, and industrial boundaries. Crucially, governments should establish integrated industry-academia-research partnerships to collectively address technical, financial, and operational challenges throughout digital infrastructure implementation, thereby ensuring its effective contribution to sustainable development goals.
Second, governments should substantially increase investment in digital infrastructure development while actively promoting the integration of intelligent management systems, IoT technologies, and big data analytics into both corporate operations and public administration frameworks. This dual approach will accelerate the digital transformation of enterprises and government entities alike, creating sustainable pathways for long-term pollution and carbon reduction. Concurrently, regional digital economy cooperation platforms should be established to facilitate intercity collaboration, enabling efficient resource sharing and complementary advantage utilization across jurisdictions. On the demand side, businesses must develop digital green consumption platforms that aggregate environmental product information and leverage big data analytics to precisely target consumers with sustainable product recommendations, thereby systematically cultivating green consumption patterns throughout society.
Furthermore, given the nonlinear diminishing returns of fixed-asset investment on digital infrastructure’s ability to synergistically enhance urban pollution reduction and carbon mitigation, governments should establish sustained monitoring and evaluation mechanisms. This mechanism should periodically assess the efficacy of digital infrastructure investments and enable dynamic adjustments to funding strategies. Once investment reaches a critical threshold, large-scale allocations toward digital infrastructure should be progressively scaled back, with priority shifting toward improving digital governance capabilities and advancing technological upgrades.
Finally, given the heterogeneous characteristics of cities, it is essential to formulate targeted policy measures customized to regional particularities. Compared with eastern regions, governments should implement targeted fiscal subsidies and policy incentives to increase digital infrastructure investment in central and western regions, thereby attracting private sector participation. Considering variations in national resource endowments, resource-rich countries should leverage their inherent advantages to drive the digital transformation of traditional industries. In contrast, resource-scarce nations can engage in regional cooperation to share resources and technologies, achieving mutually beneficial complementarity. For regions with relatively weak environmental regulations, stringent enforcement of environmental policies is imperative. The strengthened environmental oversight of enterprises, coupled with the implementation of rigorous environmental standards, can compel firms to pursue green technology innovation and industrial upgrading.

7.4. Research Limitations

This study leaves room for further exploration in several directions. First, with respect to variable measurement, while we employed the superefficiency SBM model to assess the SPCM, future research could develop more appropriate indicators for evaluating the urban SPCM. Second, in terms of mechanism analysis, the current study’s mechanism analysis does not account for structural effects; subsequent work could further investigate the structural effects. Third, concerning the research scope, our analysis did not address the spatial spillover effects of digital infrastructure, which warrants deeper examination in future studies. Finally, with respect to sample selection, extending the dataset to cross-national panel data would allow researchers to explore how digital infrastructure influences synergistic pollution–carbon governance across different types of cities in various national contexts.

Author Contributions

Conceptualization by Z.J.; methodology by Y.C.; data curation by Y.C.; formal analysis Y.C.; writing—original draft by Y.C.; writing—review and editing by Z.J.; supervision by F.Z. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Youth Project of the National Social Science Foundation of China, Grant number 21CJL028.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The data presented in this study are available upon request from the corresponding author. The data presented in this study were derived from the following resources available in the public domain: the China City Statistical Yearbook (http://www.stats.gov.cn/zs/tjwh/tjkw/tjzl/, accessed 20 July 2025); the China Statistical Yearbook on Environment (http://www.stats.gov.cn); the China Environmental Statistics Bulletin (http://www.mee.gov.cn); the China Urban Construction Statistical Yearbook (http://www.mohurd.gov.cn); and the Emissions Database for Global Atmospheric Research (https://edgar.jrc.ec.europa.eu/).

Conflicts of Interest

The authors declare that they have no conflicts of interest.

References

  1. Zhu, J.; Wu, S.; Xu, J. Synergy between pollution control and carbon reduction: China’s evidence. Energy Econ. 2023, 119, 106541. [Google Scholar] [CrossRef]
  2. Peng, H.R.; Ling, K.; Zhang, Y.J. The carbon emission reduction effect of digital infrastructure development: Evidence from the broadband China policy. J. Clean. Prod. 2024, 434, 140060. [Google Scholar] [CrossRef]
  3. Ding, C.J.; Chen, H.; Liu, Y.; Hu, J.; Hu, M.; Chen, D.; Irfan, M. Unleashing digital empowerment: Pioneering low-carbon development through the broadband China strategy. Energy 2024, 295, 131034. [Google Scholar] [CrossRef]
  4. Wei, X.; Jiang, F.; Yang, L. Does digital dividend matter in China’s green low-carbon development: Environmental impact assessment of the big data comprehensive pilot zones policy. Environ. Impact Assess. Rev. 2023, 101, 107143. [Google Scholar] [CrossRef]
  5. Zhang, H.; Wang, Y.; Chen, Y.; Wang, W. How does digital infrastructure influence synergistic effect on reducing pollution and carbon emissions? A new method based on double machine learning. J. Clean. Prod. 2025, 516, 145848. [Google Scholar] [CrossRef]
  6. Gao, D.; Li, G.; Yu, J. Does digitization improve green total factor energy efficiency? Evidence from Chinese 213 cities. Energy 2022, 247, 123395. [Google Scholar] [CrossRef]
  7. Lan, M.; Zhu, Y. Digital infrastructure construction, carbon total factor productivity, and carbon rebound effect. Environ. Sci. Pollut. Res. 2023, 30, 88968–88985. [Google Scholar] [CrossRef]
  8. Hu, J.; Zhang, H.; Irfan, M. How does digital infrastructure construction affect low-carbon development? A multidimensional interpretation of evidence from China. J. Clean. Prod. 2023, 396, 136467. [Google Scholar] [CrossRef]
  9. Li, G.; Sun, Z.; Wang, Q.; Wang, S.; Huang, K.; Zhao, N.; Di, Y.; Zhao, X.; Zhu, Z. China’s green data center development: Policies and carbon reduction technology path. Environ. Res. 2023, 231, 116248. [Google Scholar] [CrossRef] [PubMed]
  10. Zhu, Y.; Xu, Y.; Yin, S. How does digital technology innovation drive synergies for reducing pollution and carbon emissions? Sustain. Cities Soc. 2024, 116, 105932. [Google Scholar] [CrossRef]
  11. Abdullah, K.; Usman, A.M. Development of comprehensive carbon footprint and environmental impact indicators for building transportation assessment. Front. Eng. Built Environ. 2022, 2, 167–183. [Google Scholar] [CrossRef]
  12. Liang, C.; Chen, X.; Di, Q. Path to pollution and carbon reduction synergy from the perspective of the digital economy: Fresh evidence from 292 prefecture-level cities in China. Environ. Res. 2024, 252, 119050. [Google Scholar] [CrossRef]
  13. Dong, H.; Yan, Z.; Zhang, J. Does green technology innovation improve carbon emission efficiency? Evidence from energy-intensive enterprises in China. Environ. Dev. Sustain. 2024, 1–33. [Google Scholar] [CrossRef]
  14. Liu, H.; Han, P. Renewable energy development and carbon emissions: The role of electricity exchange. J. Clean. Prod. 2024, 439, 140807. [Google Scholar] [CrossRef]
  15. Yu, Y.; Zhang, N. Low-carbon city pilot and carbon emission efficiency: Quasi-experimental evidence from China. Energy Econ. 2021, 96, 105125. [Google Scholar] [CrossRef]
  16. Zhang, Y.; Ran, C. Effect of digital economy on air pollution in China? New evidence from the “National Big Data Comprehensive Pilot Area” policy. Econ. Anal. Policy 2023, 79, 986–1004. [Google Scholar] [CrossRef]
  17. Liu, B.; Qiu, Z.; Hu, L.; Hu, D.; Nai, Y. How digital transformation facilitate synergy for pollution and carbon reduction: Evidence from China. Environ. Res. 2024, 251, 118639. [Google Scholar] [CrossRef]
  18. Zhu, C.; Wang, Z.; Sun, B.; Yue, Y. Urban digital economy, environmental pollution, and resident’s health–empirical evidence from China. Front. Public Health 2023, 11, 1238670. [Google Scholar] [CrossRef]
  19. Liu, X.; Zhang, X. Industrial agglomeration, technological innovation and carbon productivity: Evidence from China. Resour. Conserv. Recycl. 2021, 166, 105330. [Google Scholar] [CrossRef]
  20. Kuzior, A.; Vyshnevskyi, O.; Trushkina, N. Assessment of the impact of digitalization on greenhouse gas emissions on the example of EU Member States. Prod. Eng. Arch. 2022, 28, 407–419. [Google Scholar] [CrossRef]
  21. Zou, W.; Pan, M. Does the construction of network infrastructure reduce environmental pollution?—Evidence from a quasinatural experiment in “Broadband China”. Environ. Sci. Pollut. Res. 2023, 30, 242–258. [Google Scholar] [CrossRef]
  22. Li, F.; Diao, Z. The impact of digital infrastructure on the synergistic effects of urban pollution and carbon reduction: Empirical evidence from China. Front. Environ. Sci. 2025, 13, 1453151. [Google Scholar] [CrossRef]
  23. Song, J.; Gao, Q.; Hu, X.; Lei, J. The impact of digital transformation of infrastructure on carbon emissions: Based on a ”local-neighborhood” perspective. PLoS ONE 2024, 19, e0307399. [Google Scholar] [CrossRef] [PubMed]
  24. Zhang, W.; Fan, H.; Zhao, Q. Seeing green: How does digital infrastructure affect carbon emission intensity? Energy Econ. 2023, 127, 107085. [Google Scholar] [CrossRef]
  25. Liu, J.; Yu, Q.; Chen, Y.; Liu, J. The impact of digital technology development on carbon emissions: A spatial effect analysis for China. Resour. Conserv. Recycl. 2022, 185, 106445. [Google Scholar] [CrossRef]
  26. Liang, J.; Qiao, C. The impact of digital economy on urban green and low-carbon development: Mechanism and spatial spillover effect. Front. Sustain. Cities 2025, 7, 1591826. [Google Scholar] [CrossRef]
  27. Mao, F.; Wei, Y.; Wang, Y. Impact of computing infrastructure on carbon emissions in China. Sci. Rep. 2024, 14, 29814. [Google Scholar] [CrossRef]
  28. Nie, J.; Shen, J.; Ren, X. Digital infrastructure, new digital infrastructure, and urban carbon emissions: Evidence from China. Atmosphere 2025, 16, 199. [Google Scholar] [CrossRef]
  29. Zhai, Z.; Zhang, T.; Yi, M.; Guan, Y.; Zhou, Y. Digital economy and the synergistic governance of pollutants and carbon emissions: Facilitation or obstruction? Environ. Res. 2024, 258, 119470. [Google Scholar] [CrossRef]
  30. Nazara, D.S.; Sutrisno, A.; Nersiwad, N.; Muslimin, M. Digital transformation in operations management: Leveraging technology to improve business efficiency. J. Oper. Manag. 2024, 1, 77–84. [Google Scholar] [CrossRef]
  31. Hakimi, M.; Amiri, G.A.; Jalalzai, S.; Darmel, F.A.; Ezam, Z. Exploring the Integration of AI and Cloud Computing: Navigating Opportunities and Overcoming Challenges. TIERS Inf. Technol. J. 2024, 5, 57–69. [Google Scholar] [CrossRef]
  32. Tan, S.; Li, W.; Liu, X.; Li, P.; Yan, L.; Liang, C. Synergistic Systems of Digitalization and Urbanization in Driving Urban Green Development: A Configurational Analysis of China’s Yellow River Basin. Systems 2025, 13, 426. [Google Scholar] [CrossRef]
  33. Zhang, J.; Liang, F.; Gao, P. Can big data reduce urban environmental pollution? Evidence from China’s digital technology experimental zone. PLoS ONE 2024, 19, e0288429. [Google Scholar] [CrossRef]
  34. Istrate, R.; Tulus, V.; Grass, R.N.; Vanbever, L.; Stark, W.J.; Guillén-Gosálbez, G. The environmental sustainability of digital content consumption. Nat. Commun. 2024, 15, 3724. [Google Scholar] [CrossRef]
  35. Niu, Y.; Lin, X.; Luo, H.; Zhang, J.; Lian, Y. Effects of digitalization on energy efficiency: Evidence from Zhejiang Province in China. Front. Energy Res. 2022, 10, 847339. [Google Scholar] [CrossRef]
  36. Sun, X. The impact of digital infrastructure on energy-environmental efficiency: Empirical evidence from China. Front. Energy Res. 2023, 11, 1277333. [Google Scholar] [CrossRef]
  37. Wang, L.; Shao, J. The energy saving effects of digital infrastructure construction: Empirical evidence from Chinese industry. Energy 2024, 294, 130778. [Google Scholar] [CrossRef]
  38. Anand, N.; Parwekar, P.; Bali, V. Energy-efficiency optimization in IoT-based machine learning for smart environmental monitoring. In International Conference on Computational Intelligence in Communications and Business Analytics; Springer: Cham, Switzerland, 2025; pp. 284–303. [Google Scholar]
  39. Wang, H.; Gu, K.; Sun, H.; Xiao, H. Reconfirmation of the symbiosis on carbon emissions and air pollution: A spatial spillover perspective. Sci. Total Environ. 2023, 858, 159906. [Google Scholar] [CrossRef]
  40. Feng, R.; Qu, X. Innovation-Driven Industrial Agglomeration Impact on Green Economic Growth in the Yellow River Basin: An Empirical Analysis. Sustainability 2023, 15, 13264. [Google Scholar] [CrossRef]
  41. Zhang, J.; Yu, H.; Zhang, K.; Zhao, L.; Fan, F. Can innovation agglomeration reduce carbon emissions? Evidence from China. Int. J. Environ. Res. Public Health 2021, 18, 382. [Google Scholar] [CrossRef] [PubMed]
  42. Wang, F. Does the construction of smart cities make cities green? Evidence from a quasi-natural experiment in China. Cities 2023, 140, 104436. [Google Scholar] [CrossRef]
  43. Han, J.; Wang, J.; Zhang, W. Digital Adoption levels and income generation in rural households in China. Heliyon 2023, 9, e21045. [Google Scholar] [CrossRef]
  44. Parrique, T.; Barth, J.; Briens, F.; Kuokkanen, A.; Spangenberg, J.H. Evidence and Arguments Against Green Growth as a Sole Strategy for Sustainability; European Environmental Bureau: Brussels, Belgium, 2019. [Google Scholar]
  45. Sanz-Torro, V.; Calafat-Marzal, C.; Guaita-Martinez, J.M.; Marti, L. An integrated approach to resource efficiency, green markets and support measures in SMEs. Int. Entrep. Manag. J. 2025, 21, 84. [Google Scholar] [CrossRef]
  46. Zhan, Y.; Ren, Y.; Xu, J. Willingness to pay a premium for eco-label products in China: A mediation model based on quality value. Sci. Rep. 2025, 15, 1783. [Google Scholar] [CrossRef]
  47. Li, C.; Wen, M.; Jiang, S.; Wang, H. Assessing the effect of urban digital infrastructure on green innovation: Mechanism identification and spatial-temporal characteristics. Humanit. Soc. Sci. Commun. 2024, 11, 320. [Google Scholar] [CrossRef]
  48. Wang, C.; Lu, W.; Yin, X.; Wang, W.; Miao, M. The Path to Green Cities: Digital Infrastructure Construction and Urban Ecological Efficiency. Pol. J. Environ. Stud. 2025, 34, 5303–5317. [Google Scholar] [CrossRef]
  49. Che, S.; Wen, L.; Wang, J. Global insights on the impact of digital infrastructure on carbon emissions: A multidimensional analysis. J. Environ. Manag. 2024, 368, 122144. [Google Scholar] [CrossRef]
  50. Cheng, W.; Ouyang, X.; Yu, A.; Shen, Z.; Vardanyan, M. Subjective perceptions versus objective outcomes: Assessing the impact of smart city pilots on environmental quality in China. Technol. Forecast. Soc. Change 2024, 209, 123799. [Google Scholar] [CrossRef]
  51. Zhao, J.; Wang, Y.; Lei, Y.; Huang, H. How does digital economy empower pollution mitigation and carbon reduction? Evidence from Chinese cities. Urban Clim. 2024, 55, 101946. [Google Scholar] [CrossRef]
  52. Jiang, T.; Liu, Y. Effect of the Digital Economy on Urban Green Development: Empirical Evidence from 284 Cities of China. Pol. J. Environ. Stud. 2024, 33, 5735–5746. [Google Scholar] [CrossRef]
  53. Tian, X.; Lu, H. Digital infrastructure and cross-regional collaborative innovation in enterprises. Financ. Res. Lett. 2023, 58, 104635. [Google Scholar] [CrossRef]
  54. Gudei, S.C.D.; Gurgu, E. The Role of IoT and Big Data Analytics in Driving Sustainable Management Transformation. Econ. Appl. Inform. 2024, 3, 36–45. [Google Scholar]
  55. Nikpour, M.; Yousefi, P.B.; Jafarzadeh, H.; Danesh, K.; Shomali, R.; Asadi, S.; Gholizadeh Lonbar, A.; Ahmadi, M. Intelligent energy management with IoT framework in smart cities using intelligent analysis: An application of machine learning methods for complex networks and systems. J. Netw. Comput. Appl. 2025, 235, 104089. [Google Scholar] [CrossRef]
  56. Guo, X.; Guo, K.; Zheng, H. Industrial agglomeration and enterprise innovation sustainability: Empirical evidence from the Chinese A-share market. Sustainability 2023, 15, 11660. [Google Scholar] [CrossRef]
  57. Liu, S.; Wu, P. The impact of high-tech industrial agglomeration on China’s green innovation efficiency: A spatial econometric analysis. Front. Environ. Sci. 2023, 11, 1167918. [Google Scholar] [CrossRef]
  58. Chen, Y.; Zhu, Z.; Cheng, S. Industrial agglomeration and haze pollution: Evidence from China. Sci. Total Environ. 2022, 845, 157392. [Google Scholar] [CrossRef]
  59. Tian, H.; Qin, J.; Cheng, C. Can industrial collaborative agglomeration improve carbon emission efficiency? Empirical evidence from China. Environ. Sci. Pollut. Res. 2023, 30, 107899–107920. [Google Scholar] [CrossRef] [PubMed]
  60. Liang, H.; Wu, Z.; Du, S. Study on the impact of environmental awareness, health consciousness, and individual basic conditions on the consumption intention of green furniture. Sustain. Futures 2024, 8, 100245. [Google Scholar] [CrossRef]
Figure 1. Geographical allocation of smart cities pilots.
Figure 1. Geographical allocation of smart cities pilots.
Sustainability 17 07066 g001
Figure 2. Analytical framework.
Figure 2. Analytical framework.
Sustainability 17 07066 g002
Figure 3. Spatiotemporal evolution of the urban reduction in pollution and carbon in China.
Figure 3. Spatiotemporal evolution of the urban reduction in pollution and carbon in China.
Sustainability 17 07066 g003
Figure 4. Para-trend test.
Figure 4. Para-trend test.
Sustainability 17 07066 g004
Figure 5. Placebo test.
Figure 5. Placebo test.
Sustainability 17 07066 g005
Figure 6. Dynamic effect of DIDM.
Figure 6. Dynamic effect of DIDM.
Sustainability 17 07066 g006
Figure 7. Single-threshold effect estimation for fix.
Figure 7. Single-threshold effect estimation for fix.
Sustainability 17 07066 g007
Table 1. Evaluation index system for the SPCM.
Table 1. Evaluation index system for the SPCM.
TypesPrimary IndicatorsSecondary Indicators
InputLaborUrban employment year-end
CapitalFixed capital stock
EnergyUrban energy consumption
Expected outputGDPEach city’s gross domestic product
Unexpected output C O 2 Each city’s annual carbon dioxide emission
Air pollutant emissions P M 2.5 concentration
Table 2. Variable definitions.
Table 2. Variable definitions.
TypesVariablesSymbolsDefinitions
Independent variableDigital infrastructureDIDVirtual variable of smart city pilot policy
Dependent variableSynergistic efficiency in pollution reduction and carbon mitigationSPCMSuperefficiency model calculation
Control variablesPopulation densitypopResident population per urban area
Educational attainmenteduEducation expenditure/general government expenditure
Environmental regulationenvComprehensive utilization rate of industrial solid waste
Mobile phone penetration ratemobileMobile subscriptions per capita
Financial development levelfinYear-end financial institution deposits/GDP
InformatizationinfPer capita telecom revenue/per capita GDP
Employment structureempNumber of tertiary industry workers/total employment
Fixed-asset investment intensityfixFixed-asset investment/GDP
Industrial structure upgradingind1 × primary sector + 2 × secondary sector + 3 × tertiary sector shares of GDP
Healthcare capacityhospHospital beds per 100 inhabitants
Technological capacityscilog of total patent applications
Mechanism variablesEnergy effectenergyGDP per unit of energy consumption
Economic agglomeration effecteaggGDP per capita
Income effectincomeDisposable income per capita
Table 3. Descriptive statistics of the main variables.
Table 3. Descriptive statistics of the main variables.
VariablesObs.MeanStd.DevMinMax
SPCM44800.2280.0990.0351.147
DID44800.2270.4190.0001.000
pop44805.7570.9141.5748.100
edu44800.1800.0420.0180.377
env448079.0623.280.240156.6
mobile44800.9790.7760.06310.17
fin44800.9850.6060.1127.450
inf44800.0390.0680.0013.372
emp448054.5713.489.91099.42
fix44800.7490.4890.008.635
ind44802.2860.1461.8312.836
hosp44800.4350.1780.0981.377
sci4480808.7613.91.0002.276
energy44804.9333.4420.00146.77
eagg44807.1161.3172.34912.00
income44801.5281.0240.16522.79
Table 4. Results of benchmark regression.
Table 4. Results of benchmark regression.
VariablesDep.Var.:SPCM
(1)(2)(3)
DID0.023 ***0.015 ***0.015 ***
(0.007)(0.006)(0.006)
Constant−0.266 ***−2.604 ***−1.549 ***
(0.074)(0.469)(0.382)
ControlsYESYESYES
Year FEsNONOYES
City FEsNOYESYES
Obs.448044804480
Adjusted R 2 0.3460.6400.704
Notes: *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively; the values in parentheses denote robust standard errors.
Table 5. Results of the PSM-DID estimators.
Table 5. Results of the PSM-DID estimators.
VariablesDep.Var.:SPCM
Nearest-Neighbor MatchingCaliper MatchingNuclear Matching
DID0.015 **0.015 **0.017 ***
(0.006)(0.006)(0.006)
Constant−1.437 ***−1.437 ***−1.364 ***
(0.344)(0.344)(0.317)
ControlsYESYESYES
Year FEsYESYESYES
City FEsYESYESYES
Obs.426342634134
Adjusted R 2 0.7060.7060.706
Notes: *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively; the values in parentheses denote the standard error.
Table 6. Results of other robustness tests.
Table 6. Results of other robustness tests.
VariablesDep.Var.:SPCM
Excluding the Impact of Other PoliciesMitigating the Impact of Nonrandom SelectionWinsorize
DID0.013 **0.014 ***0.010 **
(0.006)(0.006)(0.005)
BCS0.005
(0.006)
NBDCPZ−0.022 ***
(0.007)
Constant−1.582 ***−2.172−0.989 ***
(0.378)(1.658)(0.215)
ControlsYESYESYES
Year FEsYESYESYES
City FEsYESYESYES
City × YearNOYESNO
Obs.448044804480
Adjusted R 2 0.7070.7040.795
Notes: *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively; the values in parentheses denote the standard error.
Table 7. Results of the DML tests.
Table 7. Results of the DML tests.
VariablesDep.Var.:SPCM
Random ForestLASSO RegressionGradient LiftingVector Machine
DID0.013 ***0.013 ***0.010 ***0.015 ***
(0.003)(0.003)(0.003)(0.003)
Constant0.0000.0000.000−0.000 ***
(0.000)(0.000)(0.000)(0.000)
ControlsYESYESYESYES
Year FEsYESYESYESYES
City FEsYESYESYESYES
Obs.4480448044804480
Notes: *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively; the values in parentheses denote the standard error.
Table 8. Bacon decomposition results.
Table 8. Bacon decomposition results.
GroupingCoefficientWeight
Late T vs. Early C−0.0140.055
Treat T vs. Never C0.0110.933
Early T vs. Late C0.4500.012
Notes: Early indicates the individuals treated first, Late indicates the individuals treated later, Never indicates the individuals never treated, Treat indicates all the individuals treated, T indicates the treatment group, and C indicates the control group.
Table 9. Results of the heterogeneous treatment effect.
Table 9. Results of the heterogeneous treatment effect.
VariablesDep.Var.:SPCM
DIDMStacked DID
ATT0.016 ***0.017 **
ControlsYESYES
Year FEsYESYES
City FEsYESYES
Notes: *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively; the values in parentheses denote the standard error.
Table 10. Endogenous test.
Table 10. Endogenous test.
VariablesFirstSecond
DIDSPCM
DID 0.035 ***
(0.008)
IV2.527 ***
(4.338)
Underidentification testPassed
Kleibergen–Paap rk Wald F statistic 18.815
ControlsYESYES
Year FEsYESYES
City FEsYESYES
Obs.44804480
Adjusted R 2 0.5380.533
Notes: *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively; the values in parentheses denote the standard error.
Table 11. Mechanism test results.
Table 11. Mechanism test results.
VariablesEnergyEaggIncome
DID0.150 ***0.505 ***0.132 ***
(0.054)(0.149)(0.044)
Constant−5.168 ***−19.988 ***−15.586 **
(1.919)(5.499)(6.510)
ControlsYESYESYES
Year FEsYESYESYES
City FEsYESYESYES
Obs.448044804480
Adjusted R 2 0.7860.9000.956
Notes: *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively; the values in parentheses denote the standard error.
Table 12. Effect decomposition of the digital infrastructure and SPCM.
Table 12. Effect decomposition of the digital infrastructure and SPCM.
Transmission ChannelsDirect EffectEnergy EffectEagg EffectIncome EffectTotal Effect
Absolute contribution0.0150.00740.00860.00790.0389
Relative contribution38.56%19.02%22.11%20.31%100%
Table 13. Significance tests for threshold effects.
Table 13. Significance tests for threshold effects.
Threshold
Variables
Number of ThresholdsF Valuep ValueThreshold Value
fixSingle sill109.970.0000.285
Double sill25.650.1501.013
Table 14. Estimation results of threshold regression.
Table 14. Estimation results of threshold regression.
Threshold IntervalRegression Result
fix 0.2850.135 ***
(0.041)
fix > 0.2850.011 **
(0.005)
Adjusted R 2 0.107
Obs.4480
Notes: *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively; the values in parentheses denote the standard error.
Table 15. Heterogeneity analysis (1).
Table 15. Heterogeneity analysis (1).
VariablesDep.Var.:SPCM
Eastern RegionCentral and Western Regions
DID0.042 ***−0.001
(0.006)(0.006)
Constant−2.367 ***−0.479 ***
(0.543)(0.263)
ControlsYESYES
Year FEsYESYES
City FEsYESYES
Obs.19202560
Adjusted R 2 0.6850.758
Notes: *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively; the values in parentheses denote the standard error.
Table 16. Heterogeneity analysis (2).
Table 16. Heterogeneity analysis (2).
VariablesDep.Var.:SPCM
Resource-Based CityNon-Resource-Based City
DID0.0010.020 ***
(0.006)(0.007)
Constant−0.603−1.558 ***
(0.375)(0.479)
ControlsYESYES
Year FEsYESYES
City FEsYESYES
Obs.17922688
Adjusted R 2 0.6580.746
Notes: *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively; the values in parentheses denote the standard error.
Table 17. Heterogeneity analysis (3).
Table 17. Heterogeneity analysis (3).
VariablesDep.Var.:SPCM
Key Environmental Protection CitiesNonkey Environmental Protection Cities
DID0.020 **0.008
(0.009)(0.006)
Constant−1.826 ***−0.714 **
(0.458)(0.206)
ControlsYESYES
Year FEsYESYES
City FEsYESYES
Obs.17442736
Adjusted R 2 0.7510.707
Notes: *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively; the values in parentheses denote the standard error.
Table 18. Heterogeneity analysis (4).
Table 18. Heterogeneity analysis (4).
VariablesDep.Var.:SPCM
Two Control ZonesNoncontrolled Areas
DID0.023 ***−0.000
(0.008)(0.006)
Constant−2.084 ***−0.428
(0.520)(0.281)
ControlsYESYES
Year FEsYESYES
City FEsYESYES
Obs.24642016
Adjusted R 2 0.7430.657
Notes: *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively; the values in parentheses denote the standard error.
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

Ji, Z.; Chang, Y.; Zhou, F. Unlocking Synergies: How Digital Infrastructure Reshapes the Pollution-Carbon Reduction Nexus at the Chinese Prefecture-Level Cities. Sustainability 2025, 17, 7066. https://doi.org/10.3390/su17157066

AMA Style

Ji Z, Chang Y, Zhou F. Unlocking Synergies: How Digital Infrastructure Reshapes the Pollution-Carbon Reduction Nexus at the Chinese Prefecture-Level Cities. Sustainability. 2025; 17(15):7066. https://doi.org/10.3390/su17157066

Chicago/Turabian Style

Ji, Zhe, Yuqi Chang, and Fengxiu Zhou. 2025. "Unlocking Synergies: How Digital Infrastructure Reshapes the Pollution-Carbon Reduction Nexus at the Chinese Prefecture-Level Cities" Sustainability 17, no. 15: 7066. https://doi.org/10.3390/su17157066

APA Style

Ji, Z., Chang, Y., & Zhou, F. (2025). Unlocking Synergies: How Digital Infrastructure Reshapes the Pollution-Carbon Reduction Nexus at the Chinese Prefecture-Level Cities. Sustainability, 17(15), 7066. https://doi.org/10.3390/su17157066

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