High-quality economic growth and carbon emissions in Chinese cities: the moderating role of fiscal policies

In the rapidly evolving landscape of contemporary China, urban centers have emerged as focal points of a significant environmental challenge—carbon emissions. This comprehensive study delves into an intricate analysis, utilizing data gathered from 140 prefecture-level cities across China. Its principal aim is to dissect the effectiveness of strategies aimed at carbon reduction and fiscal policies within the multifaceted canvas of China’s urban metamorphosis, where the pursuit of high-caliber economic development takes precedence. The findings can be succinctly summarized as follows. Firstly, a statistically significant inverse correlation exists between high-quality economic development and carbon emissions in China’s urban centers. Secondly, when comparing resource-based cities to their non-resource-based counterparts, the former’s high-quality economic development plays a more prominent role in fostering carbon emission reduction. Finally, fiscal policies emerge as pivotal “accelerators” for advancing carbon emission reduction through high-quality economic development. However, their efficacy exhibits notable variations. It is essential to note that the moderating effect of environmental protection expenditures lacks statistical significance in resource-based cities and northern cities. Furthermore, the regulatory influence of resource taxation in southern cities is yet to be firmly established. This study provides practical policy recommendations for optimizing China’s eco-friendly fiscal system. These recommendations not only contribute to the realization of a green transition model for economic development but also serve as a valuable reference for governmental design of carbon emission reduction policies.


Introduction
In the realm of global sustainable development, few challenges are as critical as the specter of climate change.China, a conscientious and forward-thinking developing nation, has taken an assertive stance in the arena of global climate governance.It has earnestly pledged to reach the pivotal milestone of peak carbon dioxide emissions before 2030, followed by the ambitious quest to attain carbon neutrality by 2060.These twin objectives, collectively known as the "dual carbon" imperatives, underscore the pressing need for China to confront the formidable challenge of carbon reduction (Chen et al. 2022).Presently, China's economic landscape has undergone a transformative shift, transitioning from an era of rapid growth to one characterized by high-quality development.At the epicenter of this transformation lies the augmentation of total factor productivity, serving as the principal engine propelling high-quality economic growth.This transformation begets a crucial question: Can this paradigm shift effectively contribute to the noble cause of carbon reduction?Furthermore, the role of fiscal policies, wielding the status of potent tools for governmental intervention in both economic and environmental domains, assumes paramount significance in catalyzing high-quality economic development and the ultimate realization of the "dual carbon" objectives.It is a subject deserving meticulous attention and in-depth scholarly inquiry.
The intricate link between economic development and carbon emissions has captivated the attention of economists, environmental scientists, and governments across various echelons.A substantial corpus of research and analysis has been devoted to this subject, with a primary focus on scrutinizing the presence of the Carbon Kuznets Curve (CKC).Derived from the Environmental Kuznets Curve (EKC), which elucidates the interplay between environmental quality and per capita income (Grossman and Krueger 1994), the CKC endeavors to decipher the curvilinear connection between economic advancement and carbon emissions.Yet, a unanimous consensus on the research outcomes remains elusive.Specifically, the first set of conclusions supports the existence of the CKC (Ahmad et al. 2017;Arouri et al. 2012;Balsalobre-Lorente et al. 2022;He et al. 2021;Islam et al. 2017;Moutinho and Madaleno 2022;Vural 2020;Zambrano-Monserrate et al. 2018).However, there are differences in the turning points of the inverted U-shaped CKC.Shuai et al. (2017) empirically demonstrated that the turning point of the CKC rises with the level of economic development, indicating that the higher the level of economic development, the higher the turning point where carbon emissions transition from increasing to decreasing.The second set of conclusions lacks sufficient evidence to support the existence of the CKC (Alam et al. 2016;Bay 2020;Cheng and Da 2022;Deng et al. 2014;Du et al. 2019;Esmaeilpour Moghadam and Dehbashi 2018;Ray et al. 2023;Zoundi 2017).The third set of conclusions suggests a U-shaped or an N-shaped relationship between economic development and carbon emissions.Mrabet and Alsamara (2017) examined the case of Qatar and found a positive U-shaped relationship between carbon emissions and economic growth.Sinha et al. (2017) studied 11 emerging economies and discovered an N-shaped CKC relationship, indicating that after crossing the turning point of the inverted U-shaped curve, economic growth and carbon emissions exhibit a simultaneous increase.
The aforementioned studies have undertaken an examination of the intricate relationships between GDP growth, income levels, and carbon emissions.However, in light of China's transition into a novel phase of high-quality economic development, both governmental entities at various tiers and scholars have accentuated the paramount importance of enhancing the caliber of economic growth.Chen and Li (2022) have operationalized the concept of high-quality economic development across five pivotal dimensions: innovativeness, equilibrium, stability, sustainability, and equitability.Employing a spatial Durbin model, they have scrutinized the spatial ramifications of carbon intensity on high-quality economic development, ultimately positing a reverse causal nexus between the two variables.In a similar vein, Kuang et al. (2022), grounded in the paradigm of innovation, coordination, environmental conscientiousness, openness, and collaborative sharing, have assessed the stratum of high-quality economic development.Their inquiry extends to investigating the interplay of carbon intensity and high-quality economic development, elucidating the interrelated coupling and driving mechanisms.In a complementary vein, they have gauged high-quality economic development from the vantage points of resource utilization, economic advancement, and societal progress.They have furthermore delved into the mediating role of carbon dioxide emissions in the context of green credit, thereby elucidating the nexus between green economic development and its quality.Collectively, these scholarly endeavors converge on the assertion that the promotion of high-quality economic development significantly furthers the attainment of carbon emission reduction objectives.
Nevertheless, it is imperative to underscore that both the enhancement of economic development quality and the realization of carbon emission reduction targets are inexorably tethered to the indispensably pivotal support furnished by fiscal policies.In a recent study, Fang and Chang (2022) conducted an empirical inquiry drawing from data pertaining to the E-7 economies.Their findings substantiate the proposition that green fiscal expenditures have contributed substantively to the economic growth trajectory of Brazil.Similarly, Doğan et al. (2022) undertook a comprehensive evaluation of the impact wrought by environmental taxation and economic complexity on carbon dioxide emissions within the G7 countries.Their research underscores the efficacy of environmental taxes in effectively curtailing carbon emissions.In tandem with this, they have underscored the imperative nature of adopting decisive policy strategies, which advocate for the implementation of carbon taxation as a means to embolden G7 nations in their concerted efforts to mitigate emissions.Conversely, Ojha et al. (2020) have contended that notwithstanding the efficacy of carbon taxes in curbing carbon emissions, these levies concurrently yield a contraction in GDP.Their analysis underscores an inherent trade-off between the imperatives of economic growth and the exigencies of climate change mitigation.It is worth noting that extant scholarship predominantly gravitates towards the assessment of the impact engendered by specific tax modalities or fiscal expenditures on both economic growth and carbon emissions, largely neglecting a comprehensive investigation of the multifaceted role played by fiscal policies, the quality of economic development, and carbon emissions within an integrated research framework.
To ameliorate this conspicuous research lacuna, the present study undertakes an innovative endeavor, seeking to holistically probe into these multifarious dimensions.Our study extends the scholarly discourse in three salient dimensions: Firstly, our research aligns seamlessly with the nascent developmental epoch unfolding in China, contextualizing the enhancement of total factor productivity as the catalytic force underpinning high-quality economic development.We purport to meticulously scrutinize the impact and ramifications of this developmental trajectory on the dynamics of carbon emissions.
Secondly, our study embarks on an exploration of the pivotal role and consequential contributions made by fiscal policies in shaping the interface between highquality economic development and carbon emissions.Our aspiration is to proffer viable pathways and innovative insights that inform the reform and enhancement of green fiscal systems and policies.
Thirdly, our inquiry transcends the confines of regional and provincial data, embarking on a comprehensive investigation involving panel data sourced from 140 prefecture-level cities in China.This meticulous approach encompasses a systematic analysis carried out on a reduced spatial scale, albeit with an expanded sample size.Furthermore, our investigation discerns and distinguishes between resource-dependent municipalities and those not reliant on resource endowments, with a view to ascertaining the variances in the impact of fiscal policies across these disparate city typologies.

High-quality economic development and carbon emissions
Economic development, energy consumption, and carbon emissions are intricately interlinked.Energy, as the paramount production factor, has been the bedrock supporting China's rapid economic ascent.This trajectory of economic development, characterized by industrialization and urbanization, has, in turn, precipitated extensive energy consumption and consequential carbon emissions (Kahouli 2018).Examination of historical data encompassing China's carbon emissions and GDP growth distinctly unveils a synchronous and expeditious surge in carbon emissions, commensurate with the trajectory of economic growth, a phenomenon dating back to the inception of economic reforms.This surge in carbon emissions emerges as a concomitant outcome of rapid economic progress, thereby engendering a profound conundrum that juxtaposes environmental preservation against the ongoing pursuit of economic expansion.Within this context, China has embarked on a transformative shift in its economic development paradigm, elevating high-quality economic development to the vanguard, propelled by the amelioration of total factor productivity.Total factor productivity serves as an eminent metric, encapsulating the comprehensive efficiency of transmuting input factors into output, while concurrently dictating the intrinsic level of economic development and the potential growth trajectory.This crucial metric is fundamentally modulated by an array of factors, encompassing technological acumen, human capital, and the efficiency of resource allocation allocation (Van Beveren 2012).Building upon the judicious accumulation of human capital, the relentless augmentation of technological innovation prowess, and a discernible shift away from resource-intensive modes of production, consumption, and emissions to more efficacious alternatives, the promotion of economic and social advancement grounded in resource efficiency and environmentally sustainable development emerges as a pivotal avenue and a guarantor of the realization of carbon emission reduction objectives.
Consonant with this theoretical framework, this paper posits Hypothesis 1: China's transition into the realm of high-quality economic development, underpinned by enhancements in total factor productivity, demonstrably contributes to the reduction of carbon emissions.

High-quality economic development, fiscal policies and carbon emissions
Carbon emissions constitute externalities within the spheres of production and consumption, wherein market mechanisms inherently fall short in addressing these externalities.Consequently, it becomes imperative for the "visible hand" of government intervention to rectify this market failure (Dong et al. 2021).Fiscal policies, as the institutional bedrock and pivotal instruments of governmental regulation within the realm of economic development, occupy an irreplaceable and indispensable role in propelling high-quality economic development and carbon mitigation.Specifically, their efficacy in driving carbon reduction during the pursuit of high-quality economic development hinges fundamentally on two principal facets.Firstly, through the imposition of taxation measures, fiscal policies exert a guiding influence upon businesses and the general populace, steering them toward diminished energy and resource utilization, waste minimization, and pollutant emission reduction.These policies further engender the evolution of the economy toward low-carbon and environmentally sustainable practices (Devarajan et al. 2011).For industries and enterprises characterized by high pollution levels, corresponding environmental taxation policies are judiciously deployed, effectively incentivizing the transition from high-pollution and energy-intensive production modalities to low-carbon and environmentally benign approaches.
Secondly, fiscal expenditures channeled toward environmental preservation are judiciously employed in endeavors encompassing energy conservation, the preservation of natural forests, pollution abatement, prevention and control, land reclamation, and an array of measures expressly aimed at curtailing carbon emissions (Yuan and Pan 2023).By harnessing the synergistic effects of green taxation and green fiscal expenditure, fiscal policies effectively contribute to expediting carbon reduction throughout the course of promoting highquality economic development.
In line with this theoretical underpinning, this paper posits Hypothesis 2: Green fiscal and taxation policies serve as catalysts in accelerating the carbon reduction outcomes in the pursuit of high-quality economic development.

Research sample and data
The principal aim of this research endeavor is to systematically investigate the intricate interplay between highquality economic development in Chinese urban centers and the resultant impact on carbon emissions.In tandem with this investigation, the study endeavors to critically evaluate the effectiveness of fiscal policies within the ambit of carbon emissions reduction.The temporal scope of our study extends from the year 2011 to 2019, a time frame of paramount significance.It is pertinent to note that during this period, China's national economic and social development blueprints underwent a profound transformation.Notably, the 'Twelfth Five-Year Plan' marked a watershed moment as it incorporated explicit emission reduction targets into its binding objectives.Subsequently, the 'Thirteenth Five-Year Plan' further fortified the regulatory framework overseeing carbon emission indicators.
To empirically explore and analyze the implications of these pivotal policy developments, this research strategically selected a sample dataset drawn from a carefully considered pool of Chinese urban centers.It is widely acknowledged that these urban centers collectively represent the primary sources of carbon emissions within the nation.It is essential to underscore that, due to certain constraints related to data accessibility and the confidentiality of resource tax data in specific Chinese cities, our study draws upon a sample dataset encompassing 140 prefecture-level cities, spanning across 19 provinces.The geographical distribution of these meticulously selected cities is visually depicted in Fig. 1, thereby providing an insightful spatial representation of our study's scope.The dataset employed in this study has been meticulously curated, drawing from a diverse array of reputable sources, including the 'China Urban Statistical Yearbook, ' the 'China Financial Yearbook, ' the comprehensive statistical database maintained by WIEGO, the China Meteorological Science Data Sharing Service Platform, the Wind database, and the statistical yearbooks published by individual cities, spanning the years from 2012 to 2020.It is worth noting that in instances where data points were missing, a rigorous imputation methodology was employed, relying on data from proximate years to ensure the comprehensiveness and reliability of the dataset.

Measurement methods
The pivotal indicators underpinning this study are the levels of high-quality economic development and carbon emissions.Precise measurement and analysis of these core indicators are imperative to lay the empirical foundation for the subsequent sections of our research.

Measurement method of high-quality economic development
In the context of this study, the quantification of highquality economic development hinges significantly on the measurement of total factor productivity.This metric serves as a barometer of the overall efficiency pertaining to the conversion of inputs encompassing labor, capital, and resources into tangible outputs, thereby offering a nuanced assessment of the quality and efficiency associated with economic development.To calculate the growth rate of total factor productivity, a stochastic frontier approach (SFA) is judiciously employed as a measurement indicator.The SFA holds a distinct advantage in its capacity to delineate the production processes of individual units by estimating the production function.This approach not only facilitates the meticulous estimation of technological efficiency but also adeptly manages random errors inherent to the empirical data.The versatility of SFA renders it suitable for situations characterized by both single-input/single-output and multiple-input/ single-output scenarios, making it particularly applicable to extensive sample datasets (Wang et al. 2021).
The general form of the stochastic frontier approach (SFA) can be expressed as follows: Where i denotes the city, t the year,Y it the actual GDP, and f (•) the frontier production function, which denotes the highest output possible given a given factor input.Input factor X it consists of labor and capital stock, where labor input factor is the number of employees and capital stock is determined using the perpetual inventory technique for fixed assets.v it reflects statistical noise's random error and is an independent normal random variable with the same distribution.u it indicates the nonnega- tive random variable associated with technical efficiency, specifically the observation error and other random components that are susceptible to the γ distribution.By considering the logarithm of the production function and the derivative of time, it is possible to calculate the output growth rate, which corresponds to the total derivative of the deterministic frontier output with respect to the period t, i.e. (1) Where j = 1, 2 represent capital and labour respec- tively, and 2 j=1 ∂lnf (Xit ,t) ∂lnXitj represents the factor j of output elasticity, denoted as ∝ itj , ∂lnf (X it ,t) ∂t represents the change in technology level, denoted as TC , and − ∂u it ∂t is the derivative of technical efficiency with respect to t, denoted as the change in technical efficiency TEC.
Therefore, total factor productivity growth rate is the difference between the growth rate of output and the growth rate of input factors: Where s itj denotes the share of actual costs of element j in t time in the total costs of city i. (2)

Measurement method of carbon emissions
Chinese cities are the main source of carbon emissions.Urban carbon emissions mainly include direct carbon emissions from energy consumption, such as natural gas and liquefied petroleum gas, as well as indirect carbon emissions from the consumption of electricity and heat (Ou et al. 2010).By adding up the carbon emissions generated from the consumption of natural gas, liquefied petroleum gas, electricity, and heat, the total carbon emissions of each city can be estimated.
The carbon emissions from direct energy consumption can be calculated using the relevant conversion factors provided by IPCC 2006.It can be done by multiplying the consumption of liquefied petroleum gas and natural gas by their respective carbon emission coefficients and then summing them up.The calculation formula is as follows: Where E di denotes a certain energy i (liquefied petro- leum gas,natural gas) consumed, and θ i denotes the energy i is the coefficient of conversion to standard coal, and ǫ i denotes the carbon emission factor of the energy source i the carbon emission factor of the energy source.
The calculation of carbon emissions from electricity is mainly based on the macro-statistical method, obtained by multiplying the baseline emission factor of the Chinese power grid by the electricity consumption.The calculation formula is as follows: Where E e denotes the city's electrical energy con- sumption, and EF denotes the grid baseline emis- sion factor.In March 2022, the Ministry of Ecology and Environment in China updated the "Methods for Accounting and Reporting of Greenhouse Gas Emissions from Power Generation Facilities (Revised Edition 2022)" as an attachment to address climate change.The national grid emission factor was adjusted to 0.5810, which represents the implied carbon dioxide emissions per unit of electricity consumption.
Urban heating primarily relies on coal as a fuel source, and boiler heating and thermal power plants are the main contributors to carbon emissions from urban heat consumption in China.This study follows the approach of Wu and Guo (2016) and refers to the "GB/T 15317-2009 Monitoring of Energy Efficiency of Coal-Fired Industrial Boilers" standard, which specifies that the thermal efficiency of coal-fired industrial boilers ranges from 65 to 78%.Considering that small and medium-sized coal-fired boiler are predominant in China's centralized heating systems, a thermal efficiency (4) value of 70% is adopted.Based on the heating capacity and the heat generation coefficient of raw coal, the required quantity of raw coal can be calculated.Finally, multiplying this quantity by the conversion factor for standard coal and the carbon emission factor yields the carbon emissions from centralized heating and the carbon emissions resulting from thermal energy consumption.The calculation formula is as follows: Where E h denotes the city heat supply, NVC denotes the low-level heat output of the raw coal fuel, which is 20,908 kJ/kg, and θ c denotes the converted standard coal factor for raw coal, which is 0.7143 kg standard coal/kg, and the carbon emission factor for raw coal ǫ c is 2.53 kg CO 2 /kg.
To obtain the total carbon emissions from urban energy consumption, the carbon emissions generated from the consumption of liquefied petroleum gas, natural gas, electricity, and thermal energy are summed up.In other words, the city's total carbon emissions can be calculated as the sum of carbon emissions from: Where C denotes total urban carbon emissions and C d denotes the carbon emissions directly generated by energy consumption, and C e denotes carbon emissions from electricity consumption, and C h denotes carbon emissions from thermal energy consumption.

Dynamic characteristics of high-quality economic development
In order to delve into the dynamic attributes characterizing high-quality economic development, this study harnessed the frontier4.1 econometric software, employing the stochastic frontier approach to estimate the growth rate of total factor productivity across 140 prefecturelevel cities in China.Real GDP was considered as the output, while input factors encompassed the count of employed individuals and fixed assets (calculated using the perpetual inventory method).The specific results of these calculations are meticulously outlined in Table 7 in Appendix.Additionally, the generation of a kernel density curve was executed through Matlab software, as visually depicted in Fig. 1.This graphical representation serves to elucidate the dynamic evolutionary patterns underpinning high-quality economic development across the 140 cities.Furthermore, this analysis delves into discernible variances between cities situated in the southern and northern regions, as well as disparities between resource-based and non-resource-based cities. (6) As Fig. 2 illustrates, the kernel density function corresponding to the growth rate of total factor productivity exhibits a discernible "multi-modal" pattern.The primary peak within the density function traverses a trajectory marked by successive shifts, oscillating from left to right, and then reverting to the left once more.This intricate pattern signifies an undulating trend, characterized by an "initial decline, subsequent ascent, followed by a subsequent decline" in the overarching level of high-quality economic development experienced by the majority of cities since the initiation of the 12th Five-Year Plan.Post-2015, a distinct peak materializes on the right-hand side of the density curve, implying that select provinces have taken the lead in economic development and transitioned into a heightened realm of high-quality economic development.From the vantage point of kurtosis, it is discernible that the peaks of the kernel density curve corresponding to the growth rate of total factor productivity gradually coalesce towards the center, while the primary peak exhibits a discernible upward trajectory, denoting a narrowing trend in the high-quality economic development level among the 140 prefecture-level cities in China.
Within Fig. 2, (a1) and (a2) respectively represent the kernel density curves pertaining to total factor productivity for resource-based cities and non-resource-based cities.A comparative analysis of these two curves, (a1) and (a2), elucidates a notable degree of similarity, underscoring that the level of high-quality economic development across these two categories of cities does not exhibit marked disparities.In Fig. 2, (a3) and (a4) depict the kernel density curves characterizing total factor productivity for southern and northern cities, respectively.Upon juxtaposing (a3) and (a4), it becomes apparent that in relation to northern cities, the overall pattern of variation within southern cities is characterized by a relatively gradual trajectory, coupled with a pronounced elevation in the peak height.This signifies a converging trend within the realm of high-quality economic development levels among southern cities.

Dynamic characteristics of carbon emissions
Building upon Eqs. ( 4), ( 5), (6), and (7), we meticulously computed the total carbon emissions for each city, with detailed calculation results available in Table 8 in Appendix.Subsequently, employing Matlab software, we generated a kernel density plot to scrutinize the dynamic attributes characterizing carbon emissions across Chinese cities, as exemplified in Fig. 3.
Within the sample of 140 prefecture-level cities, the majority observed an uptick in carbon emissions from 2011 to 2019, although the extent of this growth exhibited notable variation across cities.Furthermore, the rate of growth in carbon emissions decelerated in several cities, signifying efficacious carbon reduction endeavors.As elucidated by Fig. 3, the kernel density curve pertaining to carbon emissions in Chinese cities assumes a rightskewed distribution, with the principal peak positioned on the left side, denoting a concentration of lower values.Over the observation period, the central position of the kernel density curve demonstrated minimal fluctuation, while the primary peak underwent a marginal sequence of shifts, namely a "right-shift, left-shift, and then rightshift." This culminated in an overall rightward migration, indicative of an oscillatory upward trajectory in urban carbon emissions, characterized by an "initial increase, subsequent decrease, and subsequent increase" since the 12th Five-Year Plan-a trend contrary to the alterations observed in high-quality economic development.In terms of kurtosis, the cumulative height of the primary peak in the carbon emissions kernel density curve exhibited an ascending tendency, indicative of a narrowing divide in carbon emissions among Chinese cities since the 12th Five-Year Plan.
Figure 3 delineates the (b1) and (b2) curves, which respectively represent the kernel density curves for carbon emissions in resource-based cities and nonresource-based cities. Notably, the height of the primary peak within the carbon emissions kernel density curve for resource-based cities exhibited a consistent ascent, signaling a gradual reduction in the disparity of carbon emissions among these cities.Conversely, no discernible trend toward a reduction in the carbon emissions gap among non-resource-based cities was observed.Furthermore, by examining the peaks within (b2) and (b3), it becomes apparent that the variance in carbon emissions among non-resource-based cities surpasses that among resource-based cities.Finally, (b3) and (b4) illustrate the kernel density curves for carbon emissions in southern and northern cities, respectively.These curves imply that the primary peak within the carbon emissions kernel density curve for northern cities has distinctly shifted to the right, accompanied by a reduction in height.This underscores an escalation in carbon emissions for northern cities and an expansion of the disparity.

Measurement models
Based on data from 140 prefecture-level Chinese cities going back to the 12th Five-Year Plan, this study empirically investigates the impact of China's high-quality economic development on carbon emissions.It also assesses the moderating impact of fiscal policies on the effect of high-quality economic development on reducing carbon emissions.
Firstly, a baseline regression model is constructed as follows: Where CO 2it is the explanatory variables, TFP it is the core explanatory variable, and Control it is a set of control variables, and u i , v t , ε it .The individual effects, time effects and random disturbances are represented respectively.
Furthermore, in order to examine the effectiveness of fiscal policies in moderating the carbon emission reduction effects of high-quality economic development, we integrated fiscal policies into our research framework.Building upon the baseline regression model (8), we introduced interaction terms between the level of high-quality economic development and the variables representing resource taxes and environmental protection fiscal expenditures.This allowed us to construct a moderation effects model, outlined as follows: Where Policy it are fiscal policy variables.We selected two representative variables, the proportion of resource tax revenue to total tax revenue ( RT ) and the proportion (8) of environmental protection expenditures to total fiscal expenditures ( EEP ), to serve as indicators of fiscal policies.
Core explanatory variable: the level of quality economic development ( TFP).
Moderating variable: fiscal policies ( Policy).The pro- motion of high-quality economic development and the achievement of "dual carbon" goals through fiscal and taxation policies encompass a diverse array of measures.However, due to the absence of continuous and comprehensive data, this study does not undertake a comprehensive analysis of all policy forms.Instead, it focuses on two representative variables: the ratio of resource tax revenue to total tax revenue ( RT ) and the ratio of environmental protection expenditures to total fiscal expenditures ( EEP).
Control variables ( Control ): To mitigate the potential for omitted variable bias in empirical results, four control variables were incorporated in this study, as detailed below.
Population density ( POP ): Calculated as the ratio of permanent residents to the administrative region's area.Increased population density correlates with heightened social demand and resource consumption across various categories, consequently resulting in elevated carbon emissions.
Urbanization rate ( URB ): Defined as the proportion of urban population within each city's total population.Urbanization exerts a substantial influence on urban carbon emissions, as rapid urbanization hinges on elevated energy consumption, ultimately leading to increased carbon dioxide emissions (Lin and Benjamin 2017).
Industrial structure ( IND ): Characterized by the ratio of tertiary to secondary industries.China's industrial structure remains in a stage where industrialization has not yet reached equilibrium, with a relatively high proportion of energy-intensive industries.Consequently, industrial structure stands as a pivotal factor contributing to variations in carbon emissions (Dong et al. 2020;Lu et al. 2023).
Foreign direct investment ( FDI ): Determined by the ratio of regional GDP to actual foreign direct investment utilization.Extant literature suggests that foreign direct investment (FDI) possesses the potential to either mitigate or exacerbate pollution, as it can facilitate the introduction of advanced clean and lowcarbon technologies from abroad.Conversely, it may also precipitate a "pollution haven" effect, wherein the influx of high-pollution and high-emission industries escalates carbon emissions (Liu et al. 2023;Singhania and Saini 2021;Wang et al. 2019).
Average annual temperature(TEM ): A pivotal factor influencing carbon emissions, with elevated temperatures engendering augmented demand for cooling energy, consequently increasing electricity consumption and associated carbon emissions from power generation.Furthermore, elevated temperatures can expedite chemical reactions within certain industrial processes, leading to amplified emissions of greenhouse gases and pollutants.
Table 1 presents the descriptive statistics of the specific variables under examination.

Impact of high-quality urban economic development in China on carbon emissions
Before delving into the regression results of the panel data model, it is imperative to elucidate the estimation strategy for panel data.Typically, the Hausman test is employed to discern whether a fixed effects model or a random effects model should be used for panel data model estimation.Based on the Hausman test results yielding a p-value < 0.05, thereby rejecting the null hypothesis supporting the random effects model, this study asserts that the fixed effects model is more appropriate.
Given that numerous factors influence carbon emissions, the possibility of omitting variables when selecting control variables exists.To mitigate the impact of omitted variable bias, this study adheres to the "from general to specific" modeling principle.Initially, an allencompassing regression model that incorporates all control variables is analyzed to perform a "general" empirical test.Subsequently, once the specific impact of high-quality economic development on carbon emissions is ascertained, a "specific" approach is adopted.This approach entails gradually incorporating additional control variables beyond the fundamental ones to estimate the parameters.The objective is to meticulously  scrutinize the influence of each control variable on this relationship.The specific results are presented in Table 2. Table 2 showcases the regression results at the baseline, illustrating the effect of high-quality economic growth on carbon emissions.According to Table 2, the estimation results for Models (1), ( 2), ( 3), (4), and ( 5) are obtained by progressively introducing control variables.The "general" estimation results of Model (5) reveal, with a 1% confidence level, a significantly negative coefficient for the growth rate of total factor productivity.This implies that enhancing China's level of high-quality economic development can effectively lead to a reduction in carbon emissions, thus affirming the validity of theoretical hypothesis 1.The regression results for the control variables are interpreted based on Models (1), ( 2), (3), and (4).Among these, the regression coefficients for industrial structure, population density, urbanization rate, and annual average temperature are significantly positive at the 1% level, signifying that industrial structure, population density, urbanization, and temperature rise are not conducive to achieving carbon emissions reduction, aligning with our expectations.However, the regression coefficient for the foreign direct investment ratio is negative and significantly significant at the 1% level.This implies that the influx of foreign investment can mitigate carbon emissions, suggesting that the introduction of advanced clean and low-carbon technologies from abroad has contributed to carbon emissions reduction.

Endogeneity and robustness tests
The baseline regression results reveal that an enhancement in the level of high-quality economic development leads to a reduction in carbon emissions, displaying a significant negative correlation between the two.However, it's important to consider the possibility of carbon emission reductions creating a reverse mechanism that drives high-quality economic development, establishing a bidirectional causal relationship between economic development and carbon emissions.Should the model be susceptible to potential endogeneity issues, it could yield biased or inconsistent parameter estimates.To address this concern, this study employs lagged core explanatory variables as a solution.Specifically, the lagged one-period total factor productivity serves as the core explanatory variable in the regression analysis, as depicted in the first column of Table 3.The regression coefficient remains significantly negative at a 5% level, consistent with the baseline regression results, further reinforcing the conclusions drawn from the initial analysis.This reaffirms that an improvement in the level of high-quality economic development contributes to carbon emission reductions in cities.
To bolster the robustness of the baseline regression findings, this study conducts tests that involve excluding outliers from the sample and substituting the dependent variable.Firstly, outliers in carbon emissions are rectified by winsorizing the top and bottom 1% of the sample.The results of this test are presented in the second column of Table 5.Secondly, the dependent variable is replaced with carbon intensity (carbon emissions per unit of GDP), with the corresponding test outcomes reported in the third column of Table 3. Importantly, the regression results generated through these methods consistently demonstrate the significant carbon emission reduction effect associated with high-quality economic development, underscoring the robustness of the research findings.4).The findings indicate that southern cities outperform their northern counterparts in terms of carbon emission reduction efficacy.

Evaluation of the efficacy of fiscal policies
Fiscal policies play a crucial role in promoting stable economic development, ecological governance, and environmental protection.To further examine the role of fiscal policies as "enablers" in promoting carbon emission reduction through high-quality economic development, this study investigates the moderating effect of resource taxes and environmental expenditures on the carbon reduction effect of economic development by incorporating interaction terms.The regression results, as depicted in Table 5, reveal that in column (1), based on the regression outcomes for all city samples, the coefficient of the interaction term between resource taxation and high-quality economic development is significantly negative.This signifies that resource taxation expedites the carbon emission reduction effect of high-quality economic development, thereby confirming the validity of Hypothesis 2. In columns ( 2) and (3), grounded on regression outcomes for samples from resource-based and non-resource-based cities, the efficacy of resource taxation in carbon emission reduction is evident for both types of cities. Columns ( 4) and ( 5) present the regression outcomes based on samples from northern and southern cities.The interaction term coefficients for resource taxation and high-quality economic development in northern cities are significantly negative, demonstrating a noteworthy positive regulatory role of resource taxation on the carbon emission reduction effect of high-quality economic development in these cities.However, the regulatory effect of resource taxation in southern cities has not yet been established.Based on column (1) of Table 6, it is evident that the coefficient of the interaction term between environmental protection expenditures and the level of high-quality economic development is significantly negative.This observation indicates that environmental protection expenditures expedite the carbon emission reduction effect of high-quality economic development, confirming the validity of Hypothesis 2. Columns (2) and (3) of Table 6 represent regression outcomes based on samples from resource-based and non-resource-based cities.For resource-based cities, the regression coefficient of environmental protection expenditures and its interaction term with high-quality economic development did not pass the significance test.According to columns (5) and ( 6), the interaction term coefficients for environmental protection expenditures and high-quality economic development in southern cities are significantly negative, exhibiting a significant positive regulatory effect on the carbon emission reduction effect of high-quality economic development.However, the regulatory effect of environmental protection expenditures in northern cities has not yet been established.By heeding these recommendations, policymakers can unlock the full potential of green fiscal policies, steering not only high-quality economic development but also hastening the nation's progress toward ambitious carbon reduction targets.In doing so, they lay the foundation for sustainable growth across diverse Chinese cities, aligning economic prosperity with environmental stewardship.

Fig. 1
Fig. 1 Geographical distribution of selected cities

Fig. 2
Fig. 2 Kernel density curve of total factor productivity (TFP) in Chinese cities

Fig. 3
Fig. 3 Kernel density curve of carbon emissions in Chinese cities

Table 1
Descriptive statistics of variables

Table 2
Regression results at baseline for the effect of high-quality economic growth on carbon emissions Robust standard errors for clustering to the city level are in brackets; a and c indicate significance level at the 10% and 1% respectively

Table 3
Results of endogeneity and robustness testsRobust standard errors for clustering to the city level are in brackets; a , b and c indicate significance level at the 10%, 5% and 1% respectively

Table 4
Heterogeneity analysis P-values are calculated based on the estimation results of the Chow test for the interaction model Robust standard errors for clustering to the city level are in brackets; a , b and c indicate significance level at the 10%, 5% and 1% respectively

Table 5
Moderating effect of resource taxRobust standard errors for clustering to the city level are in brackets; a , b and c indicate significance level at the 10%, 5% and 1% respectively

Table 6
Moderating effect of environmental expendituresRobust standard errors for clustering to the city level are in brackets; a , b and c indicate significance level at the 10%, 5% and 1% respectively

Table 7
Results of measuring the level of quality economic development of Chinese cities

Table 8
Results of measuring carbon emissions in Chinese cities (millions of tons)