Next Article in Journal
Impact of Climate Change on the Tourism Potential of Northeastern Brazil: Trend Analysis and Future Perspectives
Previous Article in Journal
Characterizing Drought Patterns and Vegetation Responses in Northeast China: A Multi-Temporal-Scale Analysis Using the SPI and NDVI
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Accelerating Green Growth: The Impact of Government Environmental Audits on Urban Green Economy

School of Economics and Management, Fujian Agriculture and Forestry University, Fuzhou 350002, China
*
Authors to whom correspondence should be addressed.
These authors contributed equally to the work.
Sustainability 2025, 17(12), 5289; https://doi.org/10.3390/su17125289
Submission received: 26 March 2025 / Revised: 1 June 2025 / Accepted: 6 June 2025 / Published: 7 June 2025
(This article belongs to the Section Sustainable Urban and Rural Development)

Abstract

Green growth, as a universal objective in the pursuit of sustainable development, represents a critical pathway for harmonizing economic expansion with sustainability. Within this context, government environmental auditing emerges as a pivotal mechanism for advancing the modernization of national governance systems and enhancing regulatory capacity, thereby playing an indispensable role in accelerating green transformation. This study regards green economy as a proxy variable for green development. Using panel data of cities at prefecture level and above in China from 2012 to 2021, based on the performance audit of key energy-saving and environmental protection funds conducted by the National Audit Office in 18 provinces in 2017, adopts a quasi-natural experiment method, and uses the propensity score matching double difference method (PSM-DID) to examine the impact on green development. The findings indicate that such audits significantly enhance green economy levels in audited cities. This governance instrument fosters green innovation and facilitates industrial structural optimization, reinforcing its regulatory effectiveness. Furthermore, fiscal decentralization is found to moderate the relationship between environmental performance audits and urban green economic outcomes. Additional analysis reveals that the positive impact of government environmental auditing on green economy levels is more pronounced in cities characterized by lower fiscal transparency and stricter environmental regulations. By extending the research frontier of environmental auditing through the lens of fund performance evaluation, this study offers both theoretical insights and empirical evidence to support urban green development and promote sustainable economic transitions in both developing and developed economies.

1. Introduction

The rapid advancement of industrialization has precipitated a series of environmental crises, including pollution, climate change, and resource depletion, posing severe threats to global ecosystems. Addressing these challenges has become an urgent priority worldwide. The United Nations Environment Programme (UNEP) has emphasized that environmental degradation not only undermines the sustainability of natural resources but also exerts profound impacts on economic development, public health, and social stability. In response to escalating ecological pressures, green development has become the core strategic goal of achieving sustainable transformation [1] (pp. 17–25). The green economy, as an important manifestation of this, emphasizes the pursuit of economic prosperity while taking into account ecological balance and resource efficiency and strives to establish a coordination mechanism between economic growth and environmental protection. Under the international sustainability governance framework, green development not only involves emission reduction targets and resource conservation, but also increasingly relies on government governance capabilities, institutional adaptability, and dynamic adjustments in policy implementation [2,3] (pp. 323–340, pp. 2389–2406). At present, countries around the world are striving to move away from the traditional resource-intensive development model and accelerate the transition to a green economy. Green economy has become a global development consensus. The EU’s “Green New Deal,” the United States’ clean energy investment, and the green practices of India, Brazil and other countries have jointly promoted a global move away from the resource-intensive development path to turn to an eco-friendly economic model [4,5] (pp. 101–137, pp. 373–400).
As the world’s largest carbon emitter, China plays a key role in achieving green development. According to the International Energy Agency’s 2022 CO2 Emissions Report, China’s energy-related carbon emissions reached 12.1 billion tons in 2022, accounting for 32.88% of the global total of 36.8 billion tons [6]. Given its substantial contribution to global carbon emissions, China’s transition toward a green economy is pivotal in advancing global economic sustainability. Moreover, the 2022 Environmental Performance Index (EPI), published by Yale University and Columbia University, assessed 180 countries based on key indicators such as climate resilience, environmental quality, and ecosystem vitality. China ranked 160th, underscoring the significant challenges it faces in achieving sustainable development. For the foreseeable future, fostering green economic transformation will remain a fundamental imperative for China to seek economic growth with sustainability.
In recent years, the Chinese government has implemented a series of administrative regulations and established resource conservation mechanisms to enhance resource efficiency and internalize the externalities of environmental pollution. From the perspectives of property rights and transaction costs, these measures include the enactment of the Environmental Protection Law in 2015 and the introduction of the Environmental Protection Tax in 2018—both represent significant policy measures designed to safeguard the environment and promote sustainable green development. However, the existing administrative regulations and resource conservation mechanisms mainly emphasize end-of-pipe treatment and resource management, rather than directly limiting the total amount of emissions [7] (pp. 173–188), making it difficult to fundamentally promote green transformation. In fact, in the process of promoting green development goals, the key reason why environmental pollution control has limited effectiveness lies in its strong externality characteristics and the lack of effective incentive and constraint mechanisms to encourage local governments and officials to actively fulfill their environmental governance responsibilities [8] (pp. 131–166). In this context, government environmental auditing, as an institutionalized and normalized supervision tool, has become an important mechanism to promote the effective implementation of green development goals [9] (pp. 273–274).
On the one hand, from the perspective of public trust, government environmental auditing is a form of systematic supervision and evaluation conducted by audit institutions on environmental pollution prevention and control, ecological protection policy implementation, and the use of relevant financial funds [10] (pp. 101–109), aiming to ensure that the government, as a public agent, fulfills its responsibilities in environmental governance. On the other hand, from the perspective of supervision orientation, environmental auditing can strengthen the behavioral constraints of local governments and their leading cadres in resource allocation and ecological protection through a results-oriented supervision method, thereby alleviating the problem of “lack of incentives.” Compared with the central environmental inspection system, which is mainly based on policy supervision and mainly implements the political will of superiors, government environmental auditing operates in a complete bureaucratic system and has stronger institutional independence and audit authority [11] (pp. 3–13). Its high degree of institutional autonomy can not only exert substantial pressure on the audited units, but also has a significant demonstration and deterrent effect on entities that are not directly audited, thereby encouraging more extensive compliance behaviors and promoting the institutionalization and normalization of environmental governance behaviors. This raises key questions: Can government environmental audits promote green development? What is the impact mechanism? Is there heterogeneity?
At present, some scholars have conducted research on the economic consequences of government environmental audits. From a regional perspective, research indicates that environmental auditing, through its integration with intergovernmental governance mechanisms and policy regulations, facilitates innovation in green governance strategies [12] (pp. 179–196) and significantly enhances regional environmental performance [13] pp. (1091–1106). At the corporate level, studies suggest that environmental auditing not only improves the level and quality of corporate environmental responsibility disclosures [14] (pp. 182–189) but also encourages greater corporate investment in environmental governance [15] (pp. 1271–1291), ultimately contributing to enhanced green economic performance. However, current research on government environmental audits mainly focuses on the evaluation of pollution control policies and energy conservation and emission reduction measures, and rarely involves audits of energy conservation and environmental protection funds [16] (pp. 298–311). In fact, as an important part of government environmental audits, energy conservation and environmental protection fund audits evaluate environmental effects from the perspective of funds [17] (pp. 21733–21746). Unlike general policy implementation audits, fund audits not only focus on whether environmental goals are achieved, but also focus on whether fiscal investment is reasonably allocated and whether the use of funds is efficient and transparent [18] (pp. 33–39). However, current research on its impact and its potential mechanisms is very limited. This study aims to fill this gap and provide a new perspective on the broader role of government environmental audits in promoting green economic transformation.
This study takes the performance audit of China’s key special funds for energy conservation and environmental protection as the starting point. From December 2016 to March 2017, the National Audit Office of China (NAO) conducted a comprehensive audit of the management and use of central fiscal energy conservation and environmental protection transfer funds in 2015 and 2016 in 18 provinces, autonomous regions, and municipalities directly under the central government (hereinafter referred to as provinces), and disclosed the fund management problems and rectification status of 39 prefecture-level cities in June 2017. This is the only public disclosure of the performance audit of key energy conservation and environmental protection funds since the implementation of the Interim Measures for the Management of Energy Conservation and Emission Reduction Subsidy Funds. As a concrete manifestation of the closed-loop mechanism of “Party and Government Shared Responsibilities—Five-Year Plan—Audit,” it demonstrates China’s outstanding ecological policy implementation capabilities. In addition, as the world’s largest carbon emitter and the fastest growing green economy, China has not only achieved a balance between maintaining an average annual GDP growth rate of 6% and systematically eliminating 150 million tons of steel production capacity, but has also taken the lead in incorporating “dual carbon” indicators into the audit framework, establishing a novel “policy anchoring” paradigm, and contributing valuable inspiration to global environmental governance. Based on the practice background of key energy-saving and environmental protection fund performance audits, this study uses this type of natural experiment to empirically examine the impact of government environmental audits on the level of urban green economy and the mechanism behind it.
This paper is structured as follows: Section 2 reviews literature on the green economy and government environmental auditing and proposes theoretical hypotheses. Section 3 presents the methodology and data, while Section 4 introduces and analyzes a series of empirical results. Section 5 concludes with findings from Section 4 and offers policy recommendations.
The overall research framework and analytical workflow are shown in Figure 1.

2. Literature Review and Theoretical Hypotheses

2.1. Literature Review

As a critical pathway toward sustainable development, the green economy has increasingly become a focal point of discussion in many countries. In terms of definition, existing literature suggests the green economy reflects a region’s sustainable development level and resource allocation efficiency [19] (pp. 405–424). Regarding its measurement, an increasing number of studies have adopted a multidimensional comprehensive indicator system to evaluate the degree of green economic development. Zhou et al. [20] (pp. 1–18) reviewed the application of data envelopment analysis (DEA) in energy efficiency and environmental research, providing technical support for green efficiency measurement. Song et al. [21] (pp. 361–368) emphasized that, when constructing a green economic indicator system, green innovation capabilities, resource utilization efficiency, and dynamic measurement methods driven by big data should be incorporated. In addition, some studies have also introduced a dynamic efficiency perspective, combining the Malmquist–Luenberger productivity index and the SBM model to conduct a longitudinal comparison of green economic efficiency [22] (pp. 13722–13738). Among these, the most commonly employed indicator is Green GDP, which adjusts traditional GDP by deducting environmental pollution and resource depletion [23] (pp. 187–210). However, due to challenges in obtaining precise data on pollution levels and resource depletion, cross-regional comparability remains limited. To address this limitation, scholars have shifted their focus toward green economic efficiency or green total factor productivity (GTFP). By applying input–output models and incorporating the Cobb–Douglas production function, researchers have evaluated the ability to maximize economic gains while minimizing negative environmental externalities under given regional factor endowments [22] (pp. 13722–13738). This approach provides a more objective assessment of green economic development quality.
Existing literature has explored various factors influencing green economic development, including environmental regulation, green finance, technological innovation, and public environmental awareness. Studies indicate that environmental regulation and economic openness significantly stimulate green industry growth [24] (pp. 611–619), while green technological innovation enhances resource efficiency and accelerates the transition to renewable energy [25] (pp. 65–73). Additionally, advancements in green financial instruments have diversified funding channels for the green economy [26] (pp. 1467–1482), and rising public awareness of environmental issues has further propelled market-driven sustainability initiatives [27] (pp. 68–72). These studies collectively provide a strong theoretical foundation for understanding green economic development dynamics.
Meanwhile, scholars argue that market failures stemming from environmental pollution necessitate proactive government intervention in environmental governance [28] (pp. 276–285), including environmental audits, investment in ecological protection, and the imposition of environmental taxes. Within the existing principal–agent framework between citizens and the government, government-led environmental audits serve as a key mechanism for strengthening governance effectiveness. On the one hand, government environmental auditing maintains a higher degree of independence compared with other regulatory measures, reducing risks associated with collusion in mechanisms such as environmental taxation and thereby enhancing its supervisory and attestation functions [29] (pp. 229–229). On the other hand, government environmental auditing plays a decisive role in economic oversight by closely monitoring the environmental compliance of audited entities, particularly through performance-based evaluations of fund utilization. This compensates for the limitations of traditional environmental regulatory approaches [15] (pp. 1271–1291).
Current research on government environmental auditing has primarily examined audit quality [30] (pp. 296–318), the intensity of environmental regulations [31] (pp. 9417–9438), and its micro-level effects on corporate green development and investment [15] (pp. 1271–1291). While these studies have explored the effectiveness of government environmental auditing from multiple perspectives, research focusing on the performance evaluation of environmental funds remains relatively scarce. Whether fund performance auditing can serve as an effective governance tool to incentivize urban green economic development requires further empirical investigation.
To address this gap, this study leverages a quasi-natural experiment based on China’s performance audit policy on key energy conservation and environmental protection funds. This approach enables an objective assessment of the impact and underlying mechanisms of government environmental auditing on green economic development. By uncovering the “black box” between fund performance auditing and green economic transformation, this research advances theoretical insights into environmental auditing and broadens its governance implications.

2.2. Theoretical Hypotheses

Every organization bears the responsibility of fulfilling its entrusted economic obligations [32] (pp. 75–77). Given the pervasiveness of principal–agent relationships, ensuring the effective and comprehensive execution of these responsibilities necessitates addressing agency problems such as adverse selection and moral hazard. In the context of state governance, the government operates within a unique principal–agent framework, wherein citizens entrust public resources to governmental entities. To ensure the rational and efficient allocation of these resources, robust governance structures and mechanisms must be established. This study argues that government environmental auditing serves as a critical governance instrument for advancing green development. Through its deterrence, incentive, and checks-and-balances effects, environmental auditing plays a fundamental role in strengthening the accountability framework for public-entrusted economic responsibilities.
First, through the deterrence effect, government environmental auditing enhances oversight of the allocation and utilization of industrial enterprise structural adjustment funds in audited cities. By objectively assessing the actual deployment of funds, environmental auditing identifies inefficiencies in the management of energy conservation and environmental protection funds, as well as deficiencies in policies and institutional frameworks governing key areas, such as steel and coal industry capacity reduction, new energy vehicle promotion, energy-efficient urban development, circular economy transformation parks, and building energy retrofits. By exposing these weaknesses, environmental auditing compels local governments to implement necessary corrective measures, thereby fulfilling its role in rectifying existing deficiencies [33] (pp. 64–83). Moreover, non-audited cities also experience a spillover deterrence effect, prompting them to proactively prevent regulatory violations, accelerate green transformation, and promote regional green and coordinated development. This broad enhancement of green economic performance underscores the preventive function of auditing in mitigating future environmental and governance risks [17] (pp. 21733–21746).
Second, through the incentive effect, government environmental auditing mitigates fund misallocation and environmental mismanagement by reducing information asymmetry between local governments and social investors. By fostering transparency, auditing mechanisms enhance market confidence and guide private capital toward sustainable investments, particularly in clean production and green development sectors. This, in turn, improves urban green economic performance by aligning financial flows with environmental objectives.
Finally, within a vertical governance framework, government environmental auditing functions as a checks-and-balances mechanism between central and local governments, ensuring that local authorities actively promote green development and effectively implement environmental policies. Furthermore, the public disclosure and transparency of audit findings bolster public trust in government environmental initiatives, enhance social participation, and strengthen national oversight mechanisms. This collaborative governance structure ultimately contributes to the advancement of urban green economic performance. Based on this, the following theoretical hypothesis is proposed:
H1: 
Government environmental auditing effectively enhances urban green economic performance.
Building on the theoretical foundations of compliance cost theory and the Porter hypothesis, government environmental auditing influences urban green economic performance through a dual mechanism comprising both a deterrence effect and an incentive effect. On one hand, government environmental auditing imposes stricter environmental standards and reinforces regulatory enforcement, exerting external pressure on industries regardless of their current level of environmental compliance [15] (pp. 1271–1291). This regulatory pressure compels enterprises to adopt cleaner production technologies, optimize management, and develop environmentally friendly innovations, thereby fostering green technological advancement. On the other hand, government environmental auditing also serves as a policy catalyst by guiding resource allocation and incentivizing green industries to increase investment in research and development (R&D) [12] (pp. 179–196). By enhancing green innovation capabilities, these industries drive technological progress, thereby improving resource utilization efficiency and reducing environmental pollution. The cumulative effect of these mechanisms ultimately enhances urban green economic performance. Based on this, the following theoretical hypothesis is proposed:
H2: 
Government environmental auditing promotes green innovation, generating a “technological progress” effect that positively impacts urban green economic performance.
Government environmental auditing plays a crucial role in ensuring the seamless implementation of ecological policies, bridging top-down policy design with grassroots execution through a coordinated system of checks and balances. As a result, it exerts a profound influence on urban green economic development [34] (pp. 1704–1704). On one hand, government environmental auditing enhances oversight of energy conservation and environmental protection funds, ensuring that financial resources are directed toward low-carbon, environmentally sustainable industries. This process facilitates the optimization and upgrading of high-pollution, high-energy-consuming traditional industries, increases the proportion of environmentally friendly industries, and drives the rational adjustment of urban industrial structures. On the other hand, government environmental auditing leverages information disclosure and public oversight to foster a synergistic relationship between governmental regulation and societal participation [29] (pp. 229–229). This collaborative governance mechanism reinforces the sustainable growth of green industries, enhances the efficiency and effectiveness of environmental protection initiatives, and indirectly contributes to an efficiency improvement effect. As a critical indicator of resource allocation efficiency, industrial structure rationalization optimizes factor distribution, improves resource utilization efficiency, and mitigates environmental pollution—thereby providing structural support for the advancement of urban green economies. Based on this, the following theoretical hypothesis is proposed:
H3: 
Government environmental auditing enhances urban green economic performance by optimizing industrial structures, thereby generating an “efficiency improvement” effect.
Traditional fiscal decentralization theory posits that local governments possess a deeper understanding of regional public goods demand than central authorities. Consequently, decentralized provision of public goods by local governments is often more efficient than centralized allocation. Moreover, as citizens have the right to vote on the type and quantity of public services, decision-making at lower levels of government is believed to facilitate more efficient resource allocation and equitable income distribution [35] (pp. 198–214). This decentralized governance structure not only enhances local governments’ motivation for economic development but also optimizes their capacity for resource allocation. The degree of fiscal decentralization directly influences the autonomy and resource allocation capabilities of local governments. Greater fiscal decentralization strengthens local governments’ decision-making power and financial flexibility; however, it may also exacerbate deviations in government behavior, increasing the need for regulatory oversight. In such cases, government environmental auditing plays a crucial corrective role by promptly identifying inefficiencies and ensuring the effective utilization of key energy conservation and environmental protection funds. Enhanced fiscal autonomy thus amplifies the role of environmental auditing in curbing misallocations and reinforcing green governance. Conversely, in regions with weaker fiscal decentralization, local governments have relatively lower budgetary revenues and limited financial autonomy. While environmental audits may expose inefficiencies in the use of ecological funds, financial constraints may hinder immediate rectification efforts. Although audit supervision can still partially mitigate governance inefficiencies, insufficient public funding may weaken the ability of local authorities to guide green initiatives, subsequently discouraging private sector investment in sustainable development. This financial limitation ultimately constrains the effectiveness of government environmental auditing in promoting green economic growth. Based on this, the following theoretical hypothesis is proposed:
H4: 
The relationship between government environmental auditing and urban green economic performance is positively moderated by fiscal decentralization.
The hypothesis derivation process is shown in Table 1.

3. Materials and Methods

3.1. Model Construction

This study employs the performance audit of key energy conservation and environmental protection funds conducted by the NAO in 18 provinces as a quasi-natural experiment. Using prefecture-level cities nationwide as the research sample, the cities are divided into a treatment group and a control group. Specifically, prefecture-level cities within the 18 audited provinces are classified as the treatment group, while prefecture-level cities in non-audited provinces serve as the control group. Based on the policy implementation in 2017, the study further categorizes samples into pre-policy and post-policy periods to examine the net effect of government environmental auditing. However, given the potential non-random nature of audit selection [36] (pp. 531–556), the selection of audit targets for key energy conservation and environmental protection funds may be influenced by national audit planning and resource allocation, leading to self-selection bias in audit effectiveness.
Additionally, differences in green economic performance between audited and non-audited cities may partially result from unobserved, time-invariant factors, which could introduce heterogeneity bias in direct comparisons. To address these concerns, this study first employs propensity score matching (PSM) to identify a control group comparable to the audited cities, thereby mitigating self-selection bias. Subsequently, it applies the difference-in-differences (DID) method to estimate the actual effect of government environmental auditing, ensuring result accuracy and reliability.

3.1.1. Propensity Score Matching (PSM)

Propensity score matching (PSM) is applied to construct a counterfactual by matching control group individuals to those in the treatment group based on similarity. Specifically, the sample is divided into two groups: one as the treatment group (T), indicating cities that have undergone the performance audit of key special funds for energy conservation and environmental protection, and the other as the control group (C), representing unaudited cities. Let A = {T, C} represent all sample cities. The matching process involves selecting towns from the control group (C) whose probability of receiving the performance audit of key energy conservation and environmental protection funds is highly similar to that of the audited cities. This approach helps eliminate selection bias in the estimation. Assume that the probability of a city undergoing the performance audit of key energy conservation and environmental protection funds is given by the following equation:
P = Pr A = T =     { X i , t = 1 }
where P represents the probability of a city undergoing the performance audit of key energy conservation and environmental protection funds, ∅ {·} denotes the cumulative normal distribution function, and Xi,t = 1 represents the matching variables that influence a city’s likelihood of being audited. Referring to existing studies, this study selects economic development level, industrial structure, government intervention level, digitalization level, and financial support level as matching variables. The probability formula is used to estimate the predicted probability P (X) of a city being audited. Subsequently, PSM is applied to match cities with similar predicted probabilities, thereby constructing a control group with characteristics identical to the treatment group.
Furthermore, to ensure that there is no serious multicollinearity problem between the matching variables, the variance inflation factor (VIF) test will be later performed on the covariates.

3.1.2. Difference-in-Differences (DID) Model

Based on the matched treatment and control cities, this study constructs the following model for empirical testing.
Gtfpi,t = α + β (Treati × Postt) + γ Xi,t + μi + λt + εi,t
Here, the dependent variable Gtfpi,t represents the green economy level, with i and t denoting city and year, respectively. Treati is a grouping dummy variable and Postt is a time dummy variable, with the interaction between the two being a core explanatory variable. Xi,t represents other control variables. μi denotes regional fixed effects, λt denotes time fixed effects, εt represents other random error terms, and β is the core estimation parameter.

3.2. Variable Selection

3.2.1. Dependent Variable

In this study, although “Green Growth” and “Green Economy” are closely related, they have different focuses. Green growth emphasizes minimizing environmental pressure while achieving economic growth goals, and emphasizes the sustainability of the “growth path;” while green economy focuses more on the structural transformation of the economic system, and realizes the greening of the “economic structure” through the development of green industries, the improvement of resource efficiency and the protection of the ecological environment.
The green economy reflects the comprehensive efficiency of economic growth and environmental pollution control. Currently, academia primarily uses green GDP, green economic efficiency, and GTFP to characterize the green economy level [37,38] (pp. 1–35, pp. 3163–3185). Among these, green GDP mainly deducts the expenses associated with resource depletion and environmental degradation from GDP, but its complex calculation is error prone. Green economic efficiency only selects CO2 as the undesirable output. In contrast, the measurement indicators of GTFP not only incorporate regional inputs and desirable outputs but also include undesirable outputs such as COD, SO2, ammonia nitrogen, and nitrogen oxides, offering a more reliable measure of regional green economic levels. Therefore, following the research of Bigsten et al. [39] (pp. 1423–1441), this study selects the number of urban employees (in tens of thousands), fixed asset investment adjusted by the perpetual inventory method (in tens of thousands of CNY), and urban energy emissions (social electricity consumption) as input indicators. The actual GDP of the city is used as the desirable output indicator, while industrial SO2 emissions, industrial wastewater emissions, and industrial dust emissions are considered undesirable outputs. By combining the data envelopment analysis (DEA) method; the non-radial, non-angular slack-based measure (SBM) efficiency model for undesirable outputs and slack problems; and the Malmquist–Luenberger productivity index, the urban green economy level is calculated.

3.2.2. Independent Variables

The explanatory variables in this study include the grouping dummy variable (Treati), the time dummy variable (Postt), and their interaction term (Treati × Postt). Specifically, Treati indicates whether the National Audit Office has audited a city for the key special funds on energy conservation and environmental protection. As the National Audit Office conducted the audit on special funds for energy conservation and environmental protection from December 2016 to March 2017, the year 2017 is set as the policy shock point, with Postt = 1 after this period and Postt = 0 otherwise. The interaction term Treati × Postt is the core explanatory variable, and its coefficient is the key parameter of interest to be estimated.

3.2.3. Control Variables

In addition to government environmental auditing, other factors may also influence urban green economic performance. Referring to existing studies [40,41] (pp. 23–36, pp. 106–114), this study selects the five following control variables:
(1)
Economic Development Level: Measured by the logarithm of each city’s GDP. Cities with higher economic development levels tend to allocate more resources to green technology innovation and environmental protection.
(2)
Industrial Structure: Measured by the ratio of tertiary industry-added value to secondary industry-added value in each city. Since tertiary industries typically consume less energy and generate lower emissions than secondary industries, an increase in the share of tertiary sectors is expected to promote green economic development.
(3)
Government Intervention Level: Measured by the ratio of general budgetary fiscal expenditure to GDP in each city. Government fiscal spending supports environmental protection, green technology innovation, and infrastructure development, facilitating the transition to a green economy.
(4)
Digitalization Level: Measured by the “Peking University Digital Financial Inclusion Index” (divided by 100). Digital technologies can improve resource utilization efficiency and drive the transformation of traditional industries toward green and low-carbon development.
(5)
Financial Support Level: Measured by the ratio of year-end financial loans to GDP in each city. Financial resources provide funding for green projects, fostering green technology innovation and industry growth.

3.2.4. Mechanism Variables

Based on the analysis in this study, the mechanism variables through which government environmental audit may affect urban green economy levels include green innovation, industrial structure rationalization, and fiscal decentralization. In brief, green innovation level is measured by the number of green innovation patents granted per 10,000 people; industrial structure rationalization is measured using the Theil index; and fiscal decentralization is quantified by the ratio of general fiscal revenue to expenditure of city governments.

3.3. Data Sources

This study utilizes panel data from Chinese cities spanning 2012 to 2021 to examine the impact of government environmental auditing on urban green economic performance. Specifically, data on GTFP inputs and outputs are obtained from the China Urban Statistical Yearbook, China Regional Statistical Yearbook, China Energy Statistical Yearbook, and China Environmental Statistical Yearbook. Other variable data are sourced from the China Urban Statistical Yearbook and the Peking University Digital Finance Research Center’s official website. For missing data, interpolation methods and autoregressive integrated moving average (ARIMA) models are applied for estimation and prediction. Table 2 summarizes the names and calculation methods of the main variables.
Among the data used in this study, the calculation of the Green Economic Efficiency Index (GTFP) was completed independently by the research team based on the original statistical data, including the integration of input and output variables, the setting of the SBM model, the compilation of the GML index and other technical links. The values of variables such as green innovation patent data (Lscx), fiscal decentralization (Czfq), and industrial structure index (Thei) are from official or third-party statistical databases, and the variable construction method is independently set by the author. In addition, some basic data, such as urban GDP, industrial structure, and fiscal expenditure, are borrowed from existing data in authoritative publications such as the China City Statistical Yearbook, China Energy Yearbook, and China Environment Yearbook. Although these data are public statistical results, they are still screened by the research team, processed in a unified caliber, and missing values are interpolated during use.

4. Results

4.1. Model Testing Based on PSM-DID

4.1.1. VIF Test

To account for potential multicollinearity among variables, this paper conducted a variance inflation factor (VIF) test on the key independent variables (Treati × Postt) and all control variables (Eco, Con, Fis, Dig, Size) in the model to identify multicollinearity. The results are presented in Table 3. The overall mean VIF value is 1.85, and all individual VIF values are below 5, indicating no significant multicollinearity issue among the selected variables that could affect regression robustness.

4.1.2. Propensity Score Matching (PSM)

Following Becker and Ichino [42] (pp. 358–377) and Liu et al. [43] (pp. 63–75), this study applies the nearest neighbor matching within a caliper method. Six covariates are selected for matching between the treatment group and control group, including green economy (Gtfp), economic development level (Eco), industrial structure level (Con), government intervention level (Fis), digitalization level (Dig), and financial support level (Size). To ensure comparability, samples that do not satisfy the common support assumption are removed, thereby reducing systematic differences between the treatment and control groups after matching. The matching results are presented in Table 4. Before matching, key covariates showed substantial biases, industrial structure and digitalization had standardized differences exceeding 60% and 90%, respectively. After matching, all covariates standardized biases fell below 10%, with most under 5%, indicating that PSM effectively improved group comparability and reduced selection bias. The model shows that a total of 2473 samples satisfy the common support assumption. Additionally, the absolute value of standardized deviations after matching is below 10%, indicating that the matched treatment and control groups have similar characteristics after PSM and pass the balance test. Variable descriptive statistics are shown in Table 5.
The dependent variable Gtfp has a mean of 1.0076, a standard deviation of 0.0501, a minimum value of 0.3734, and a maximum value of 1.5075. This indicates that green economic efficiency exhibits relatively tiny fluctuations, with values concentrated around 1.0. The interaction variable Treati × Postt has a mean of 0.3744 and a standard deviation of 0.4841. This suggests that this variable follows a binary distribution in the sample, with a relatively even distribution of 0 s and 1 s.

4.1.3. Difference-in-Differences (DID) Test

The difference-in-differences (DID) model was employed to identify the net effect of government environmental audits on the level of urban green economy. The relevant results are presented in Table 6. Specifically, column (1) shows the regression results without control variables but with city fixed effects and time fixed effects. Column (2) builds on column (1) by adding the control for regional economic development level. Column (3) further extends column (2) by including controls for economic development level, industrial structure level, government intervention level, digitalization level, and financial support level. The results indicate that the regression coefficients of the core explanatory variable, Treati × Postt, are significantly positive at the 1% level. Compared with cities without audits, cities subjected to government environmental audits experienced an average increase of 1.12% in their green economy levels. This finding remains significant after incorporating a series of control variables, confirming hypothesis 1. Specifically, government environmental audits effectively fulfill the functions of “identifying problems,” “Treating existing issues,” and “preventing potential risks,” thereby forming a synergistic oversight mechanism. This mechanism effectively promotes urban energy conservation, emission reduction, and green innovation, ultimately enhancing the level of urban green economy. This is evidenced by the case of the Quanzhou City Audit Bureau, which, under the guidance of government environmental audits, has intensified its audit efforts in areas such as carbon sequestration, energy conservation, emission reduction, and green development. These efforts have contributed to low-carbon development and the transformation of green growth, resulting in a notable improvement in the city’s green economy level. Compared with existing studies, this research finds that performance audits of funds have significant supervisory and corrective effects [29] (pp. 229–229), thereby broadening the research perspective on government environmental audits to some extent.

4.2. Parallel Trend and Dynamic Effect Test

The effectiveness of the difference-in-differences (DID) method in policy evaluation relies on the parallel trend assumption, which requires that, without the performance audit of key energy conservation and environmental protection funds, the treated cities should exhibit the same trend as the control group. To verify this assumption, following the methodologies of Roth [44] (pp. 305–322), the following model is constructed:
G t f p i , t = α + β t 2013 2021 T r e a t i × Y e a r t + γ X i , t + μ i + λ t + ε i , t
In this model, Yeart represents the year dummy variable, while other variables are consistent with Model (1). The first year of the sample period (2012) is set as the benchmark year, and no interaction term between Treati and Yeart is included for that year. The key coefficient of interest, βt, represents the difference in green economic performance between treated cities and control cities for each year t (t = 2013, 2014, …, 2021). Figure 2 presents the estimated coefficients βt, with dashed lines indicating the 95% confidence interval. The results indicate that, from 2013 to 2016, the regression coefficients remained close to zero, suggesting no significant difference in green economic performance between the treatment and control groups before the performance audit of key energy conservation and environmental protection funds. Although there was a slight decrease in 2016, possibly due to China’s proposal to levy an environmental protection tax, the overall trend confirms that the parallel trend assumption is satisfied. After the implementation of the audit in 2017 and 2018, the regression coefficients became significantly positive, indicating that government environmental auditing effectively enhanced urban green economic performance. However, in 2019 and 2020, the estimated coefficients were insignificant, suggesting that the deterrent effect of a single audit is not sufficient [40] (pp. 23–36). This finding highlights the need for regular and routine audits by relevant government agencies to improve the effectiveness of environmental auditing further.

4.3. Robustness Tests

4.3.1. Outlier Test

Considering the presence of extreme values in prefecture-level city variables, which may affect the baseline regression results, it is necessary to conduct winsorization on continuous variables. Therefore, a 1% two-tailed winsorization is applied to the following variables: GTFP, economic development level, industrial structure level, government intervention level, digitalization level, and financial support level. The results are presented in column (1) of Table 6. The regression findings indicate that, after applying a 1% two-tailed winsorization, the positive effect of government environmental auditing on urban green economic performance remains significant. Furthermore, the significance level improves, providing additional support for hypothesis 1.

4.3.2. Replacing the Dependent Variable

In the baseline regression, urban green economic performance was measured using the non-radial, non-angular SBM-GML index, which incorporates undesirable outputs and slacks. To test robustness, this study replaces the dependent variable by measuring green economic performance using the non-radial, non-oriented SBM-DDF model and re-estimates model (3).
The results, presented in column (2) of Table 7, indicate that, after replacing the dependent variable, the positive effect of government environmental auditing on urban green economic performance remains significant at the 5% level. This further validates hypothesis 1.

4.3.3. Placebo Test

The omission of essential control variables may lead to biased baseline estimates, making it essential to test whether omitted variables affect the results. Following the approach of Peisakhin and Rozenas [45] (pp. 535–550), this study conducts a placebo test by randomly selecting 300 samples from the 2473 observations across 30 provinces from 2012 to 2021 as a pseudo treatment group, while the remaining 2173 samples serve as a pseudo control group. This process is repeated 1000 times to examine whether the baseline results change. Because groups are randomly assigned, the average treatment effect should center around zero. As shown in Figure 3, the density probability distribution of the estimated coefficients from 1000 randomly generated samples indicates that most of the coefficient estimates are concentrated around zero, indicating that the baseline regression results pass the placebo test and that the findings are not significantly biased by omitted variables.

4.3.4. Shortening the Sample Period

Apart from the government environmental auditing event, other random factors may also influence urban green economic performance. To eliminate potential external influences, this study follows the approach of Chen et al. [46] (pp. 99–122) by shortening the sample period. Specifically, the sample is restricted to three years before and after the audit event, retaining city-level data from 2014 to 2019 for estimation. The results in Columns (1) and (2) of Table 7 show that the coefficient of Treati × Postt remains significantly positive at the 1% level, confirming the robustness of the baseline results regardless of sample period selection.

4.3.5. Double Machine Learning Model Replacement

Considering the nonlinear relationships among variables in the process of green economic growth, conventional linear model estimations may lack robustness. Following the approach of Zhang and Li [47] (pp. 113–135), this study employs a partially linear double machine learning model to estimate the impact of government environmental auditing on urban green economic performance.
Gtfpit + 1 = β (Treati × Postt) + g(Xit) + Uit
E (Uit|Treati × Postt, Xit) = 0
where Xit represents a set of high-dimensional control variables, whose specific functional form ĝ(Xit) is estimated using machine learning algorithms. Uit is the error term, with a conditional mean of zero, while the definitions of other variables remain unchanged.
By estimating Equations (4) and (5), the key coefficient of interest can be obtained. The double machine learning results in column (3) of Table 8 indicate that the coefficient of Treati × Postt is 0.0114, significantly positive at the 1% level. The estimated correlation coefficient also exhibits a substantial improvement, further validating the robustness of hypothesis 1.

4.3.6. Excluding the Impact of Other Policies

In June 2017, the People’s Bank of China (PBOC), along with the China Banking Regulatory Commission (CBRC) and five other ministries, jointly issued the Guiding Opinions on Establishing a Green Financial System in Five Green Finance Reform and Innovation Pilot Zones. The policy designated Ganjiang New District in Jiangxi Province; Huzhou and Quzhou in Zhejiang Province; Huadu District in Guangzhou (Guangdong Province); Gui’an New District in Guizhou Province; Hami City in Changji Prefecture; and Karamay City in Xinjiang as the first batch of financial reform and innovation pilot zones. Because green finance pilot zones may influence urban green economic development, this study removes the sample cities that were both audited and located in the green finance reform pilot zones—specifically, cities in Zhejiang and Guangdong Provinces—and re-estimates the baseline model (1). The regression results in columns (1) and (2) of Table 9, show that, after excluding the impact of green finance reform pilot zones, the coefficient of the key explanatory variable increases further. Even after including control variables, time fixed effects, and regional fixed effects, the coefficient remains significantly positive at the 1% level. This confirms that the green finance reform pilot zone policy does not influence the effectiveness of government environmental auditing when enhancing urban green economic performance.

4.4. Mechanism Test

The baseline estimation results indicate that government environmental auditing effectively enhances urban green economic performance. Theoretically, there are two primary channels through which green economic development improves: technological progress and efficiency improvement [48] (pp. 57–69). The question remains: Can the deterrence, incentive, and checks-and-balance effects of auditing be realized through technological progress and efficiency improvement? Therefore, it is necessary to empirically examine the specific pathways through which government environmental auditing enhances urban green economic performance. Following the approach of Jiang [49] (pp. 100–120), this study constructs the following mediation effect model to investigate the underlying mechanisms.
Mi,t = α + β (Treati × Postt) + γ Xi,t + μi + λt + εi,t
In the equation, Mi,t represents the mediating variable, which sequentially refers to urban green innovation (Lscx) and industrial structure rationalization (Thei) in the subsequent analysis. Other variables are defined as in model (2).

4.4.1. Testing the “Technological Progress” Pathway

The previous analysis indicates that green innovation, a key manifestation of technological progress, is crucial for promoting green economic development. To verify whether green innovation mediates the relationship between government environmental auditing and urban green economic performance, this study follows the approach of Pei Mengdi [50] (pp. 16–26). The number of green innovation patents granted per 10,000 people is used as a proxy variable for green innovation, and model (6) is estimated accordingly. The mechanism test results in column (1) of Table 8 show that the interaction term’s regression coefficient remains significantly positive at the 1% level. This confirms that government environmental auditing enhances urban green economic performance by promoting green innovation, thereby demonstrating a “technological progress” effect and validating hypothesis 2.

4.4.2. Testing the “Efficiency Improvement” Pathway

Industrial structure rationalization is a key approach to enhancing resource allocation efficiency. It may serve as a critical mediating factor in the relationship between government environmental auditing and urban green economic performance. To examine whether industrial structure rationalization mediates the impact of government environmental auditing on urban green economic performance, this study follows the approach of Dong et al. [51] (pp. 468–477). The Theil index is used to measure the level of industrial structure rationalization, and model (5) is estimated accordingly. The mechanism test results, presented in column (2) of Table 10, indicate that the interaction term’s regression coefficient remains significantly positive at the 1% level. This suggests that government environmental auditing effectively promotes industrial structure rationalization, thereby demonstrating an “efficiency improvement” effect. Consequently, hypothesis 3 is supported.
The logical framework is shown in Figure 4.

4.5. Moderating Effect Test

As discussed in the theoretical analysis, fiscal decentralization may moderate the effect of government environmental auditing. Following Jia et al. [52] (pp. 107–122), this study measures fiscal decentralization intensity using the ratio of general fiscal expenditure to revenue and constructs a moderating effect model for empirical testing.
GTFPi,t = α + β (Treati × Postt) + θCzfq + g (Czfq × Treati × Postt)+ γ Xi,t + μi + λt + εi,t
In the equation, Czfq represents the moderating variable (fiscal decentralization), and Czfq × Treati × Postt is the interaction term between the moderating variable and the explanatory variable. θ denotes the coefficient of the moderating variable, while g is the key parameter to be estimated. The definitions of other variables remain consistent with those in the previous sections.
The results in Table 11 indicate the following: Column (1) confirms that government environmental auditing has a significant positive impact on urban green economic performance; column (2) shows that the interaction term between fiscal decentralization and the audit of key energy conservation and environmental protection funds is significantly positive. This suggests that fiscal decentralization strengthens the relationship between government environmental auditing and urban green economic performance—that is, higher fiscal decentralization enhances the impact of government environmental auditing on green economic development. Empirical evidence supports this finding. Greater fiscal decentralization grants local governments more autonomy in taxation, expenditure, and budgeting, which in turn intensifies their influence over resource allocation. In this context, government environmental auditing can flexibly optimize resource distribution, promptly identify issues, and ensure the effective utilization of key energy conservation and environmental protection funds. Additionally, the corrective function of auditing becomes more pronounced, reinforcing its role in enhancing urban green economic performance.

4.6. Heterogeneity Analysis

4.6.1. Fiscal Transparency

Fiscal transparency is a key mechanism for enhancing resource allocation efficiency and ensuring the accountability of governments and public officials. However, it remains uncertain whether the effectiveness of government environmental auditing varies under different levels of fiscal transparency. To explore this heterogeneity, this study follows the “China Municipal Government Fiscal Transparency Research Report,” compiled by the 21st Century Development Institute and the Public Economics, Finance, and Governance Research Center, to measure local government fiscal transparency. Cities are divided into high and low fiscal transparency groups based on the median fiscal transparency. The grouped regression results in columns (1) and (2) of Table 10, show the following: In cities with lower fiscal transparency, the coefficient of Treati × Postt is significantly positive at the 1% level and in the towns with higher fiscal transparency, the coefficient of Treati × Postt is insignificant. These findings suggest that, in cities with lower fiscal transparency, the deterrent effect of the National Audit Office’s performance audit on key energy conservation and environmental protection funds is more substantial, leading to a more significant improvement in green economic performance. Conversely, in cities with higher fiscal transparency, the performance audit fails to exert a strong deterrent effect, resulting in insignificant improvements in green economic performance. One possible explanation is that cities with higher fiscal transparency already have well-established budget disclosure mechanisms, comprehensive fiscal data reporting, transparent decision-making processes, and robust fiscal oversight and audit mechanisms. As a result, their green economic performance is already relatively high, making the marginal impact of the audit less pronounced.

4.6.2. Environmental Regulation

Given that the impact of government environmental auditing on urban green economic efficiency may vary across regions with different levels of environmental regulation, this study follows the approach of Guo and Wang [53] (pp. 2117–2129). The urban, comprehensive utilization rate of general industrial solid waste is used as a proxy for environmental regulation levels, and cities are divided into low and high environmental regulation groups based on the median level of environmental regulation. As shown in columns (3) and (4) of Table 12, the coefficient of Treati × Postt is insignificant in regions with lower environmental regulation but significantly positive at the 1% level in areas with higher environmental regulation. These findings suggest that, in cities with weaker environmental regulations, government environmental auditing does not considerably influence urban green economic performance. One possible explanation is that local governments in these regions do not prioritize environmental protection, leading to a weaker impact of environmental auditing on green economic performance. Conversely, in cities with stricter environmental regulations, local governments place greater emphasis on environmental protection. As a result, issues identified through auditing are more effectively rectified, allowing government environmental auditing to enhance urban green economic performance significantly.

5. Discussion

This study theoretically expands the research framework of the impact of government environmental audits on environmental governance and green economic performance, as follows:
First, this paper confirms the effectiveness of government environmental audits in improving urban green economic performance, expanding the boundaries of green economy and government audit research. Unlike Wu et al. [54] (pp. 1213–1241) and Zhang [55] (pp. 813–844), who focus mainly on the impact of audits on environmental governance and resource utilization, this paper extends the research perspective to its driving role in green economic development. By constructing the PSM-DID model, this paper fills the gap in empirical research on the impact of government environmental audits on green economic performance, especially in revealing a path by which to improve urban green economic efficiency through green innovation and industrial structure optimization, providing a new perspective for green economic theory.
Secondly, we identify the internal mechanism of the government environmental audit in improving green economic performance. Compared with existing research that mainly focuses on the effectiveness of environmental policy implementation [34] (pp. 1704–1704), this study focuses on the systematic role of government environmental auditing in promoting technological progress, improving the efficiency of capital use and optimizing industrial structure, and reveals its core function in the sustainable development of green economy from the perspective of resource allocation and performance orientation. This finding provides a solid foundation for deepening the theoretical discussion on the interactive mechanism between government audit and green development.
Third, our analysis also found, in its heterogeneity analysis, that the effects of government environmental audits in different cities are significantly different, especially in areas with low fiscal transparency and high environmental regulation intensity. This result is not only consistent with the research conclusion of Xu Jianjun et al. [56] (pp. 86–92) on the impact of government environmental audits on green economic performance, but also echoes the OECD’s discussion on the relationship between fiscal transparency and policy effectiveness, as well as the logic of environmental regulation promoting innovation and compensation effects proposed in the “Porter hypothesis” [57] (pp. 97–118). In summary, the above results show that the policy effect of government environmental audits has obvious contextual dependence. In order to improve the accuracy and effectiveness of green economic policies, policymakers should optimize the design and implementation of audit systems in accordance with local conditions based on the specific characteristics of cities in terms of fiscal governance and environmental regulation.
Although this study has made useful explorations in the relationship between government environmental audits and urban green economy, it still has certain limitations, which in turn also provide directions for subsequent research.
First, at the data level, this article mainly relies on public data at the prefecture or city level, and has not yet deeply examined micro-subjects such as enterprises. As the core unit of market activities, the behavior mechanism of enterprises may be significantly different from the green economic performance at the city level.
Second, in terms of mechanism analysis, this article mainly starts from the perspective of technological progress, efficiency improvement and fiscal decentralization, but lacks a more detailed identification of the evolution path and operation dynamics of the audit role.
Third, there are still omissions in the heterogeneity analysis, and other regional characteristic variables that may affect the development of the green economy have not been fully included. Subsequent research can further expand the analysis dimension.
Finally, there is still room for improvement in policy recommendations. The current recommendations mainly focus on the overall direction, and, in the future, policy tools should be further refined to enhance their pertinence and operability, so as to optimize the existing audit and green development policy system.

6. Conclusions

This paper takes the performance audit of energy-saving and environmental protection funds as the starting point, deeply analyzes the impact of government environmental audits on the development of urban green economy, and draws several important conclusions.
First, the empirical results show that government environmental auditing significantly improves the level of urban green economy, and this conclusion still holds after a series of robustness tests.
Second, the mechanism analysis finds that government environmental auditing effectively improves urban green economy by promoting green innovation and optimizing industrial structure.
Finally, the heterogeneity analysis shows that, in cities with low fiscal transparency and high environmental regulation level, the positive impact of government environmental audit on urban green economic performance is more significant.
Based on the above analysis, this study puts forward the following policy recommendations:
First, the authority of audit work should be enhanced. Audit departments should appropriately expand the coverage of special audits and strengthen the tracking and results disclosure of key projects, so as to enhance the binding force of an audit on risk prevention and control and policy implementation.
Second, the deterrent effect of audit work should be deepened. Government departments should innovate the form of information disclosure, flexibly use online media, broaden the channels for information disclosure, guide the public to participate in supervision, improve the transparency of fiscal information, and build a sunshine government.
Third, multiple measures should be taken to improve the degree of fiscal decentralization. Government departments should optimize the expenditure structure, strengthen performance management and improve the business environment, guide enterprises to invest and start businesses, and increase fiscal revenue.
Then, we should promote the deep integration of the environmental audit system with the international sustainable governance framework. In view of the global common characteristics of green development, the government environmental audit system should be effectively connected with the United Nations Sustainable Development Goals (SDGs), and the system should include “governance capacity improvement,” “system adaptability construction” and “long-term sustainable performance” evaluation indicators.
Finally, an audit problem rectification list system should be implemented. Government departments should clarify the responsible units for rectification, refine the rectification measures, implement the rectification results, and realize the closed-loop management of audit problems through information systems. In addition, it is necessary to strengthen the coordination and linkage between the National Audit Office, the Finance Department, and the Taxation Department, so that they may work together to promote problem rectification and strive to open up the “last mile” of audit work.
Given that this study still has limitations in terms of data level, mechanism identification and variable control, enterprise-level data and dynamic methods can be introduced in the future to further reveal the micro-path and time-effect characteristics of the audit effect. At the same time, heterogeneous variables and policy simulation analysis should be expanded to improve the accuracy and empirical support of government environmental audit intervention.

Author Contributions

Conceptualization, X.L. and B.D.; methodology, X.L. and B.D.; software, X.L. and B.D.; validation, X.L., B.D. and S.L.; formal analysis, X.L. and B.D.; investigation, S.L.; resources, B.X.; data curation, L.X.; writing—original draft preparation, X.L. and B.D.; writing—review and editing, X.L. and B.D.; visualization, B.X. and L.X.; supervision, B.X. and L.X. All authors have read and agreed to the published version of the manuscript.

Funding

National Social Science Foundation of China (19BGL087).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

References

  1. Keeble, B.R. The Brundtland report: ‘Our common future’. Med. War 1988, 4, 17–25. [Google Scholar] [CrossRef]
  2. Meadowcroft, J. What about the politics? Sustainable development, transition management, and long term energy transitions. Policy Sci. 2009, 42, 323–340. [Google Scholar] [CrossRef]
  3. Wu, J.; Guo, S.; Huang, H.; Liu, W.; Xiang, Y. Information and Communications Technologies for Sustainable Development Goals: State-of-the-Art, Needs and Perspectives. IEEE Commun. Surv. Tutor. 2018, 20, 2389–2406. [Google Scholar] [CrossRef]
  4. Spring, C.R.; Cirella, G.T. Fostering sustainable development: Green energy policy in the European Union and the United States. In Human Settlements: Urbanization, Smart Sector Development, and Future Outlook; Springer Nature: Berlin/Heidelberg, Germany, 2022; pp. 101–137. [Google Scholar]
  5. Letti, L.A.J.; Sydney, E.B.; de Carvalho, J.C.; de Souza Vandenberghe, L.P.; Karp, S.G.; Woiciechowski, A.L.; Soccol, V.T.; Novak, A.C.; Junior, A.I.M.; Burgos, W.J.M.; et al. Roles and impacts of bioethanol and biodiesel on climate change mitigation. In Biomass, Biofuels, Biochemicals; Elsevier: Amsterdam, The Netherlands, 2022; pp. 373–400. [Google Scholar]
  6. Dong, B.; Chen, L.; Zhang, Y.; Xie, B. How does green credit supply enhance the efficiency of forestry ecological development? Taking the perspective of ecological total factor productivity. J. Clean. Prod. 2024, 488, 144643. [Google Scholar] [CrossRef]
  7. Wang, X.; Wang, Y. Research on green credit policy to promote green innovation. Manag. World 2021, 37, 173–188. (In Chinese) [Google Scholar]
  8. Acemoglu, D.; Aghion, P.; Bursztyn, L.; Hemous, D. The Environment and Directed Technical Change. Am. Econ. Rev. 2012, 102, 131–166. [Google Scholar] [CrossRef]
  9. Nazarova, K.; Nezhyva, M.; Kucher, A.; Hotsuliak, V.; Melnyk, T.; Zaremba, O. Environmental Audit in the Sustainable Development of Green Economy. Eur. J. Sustain. Dev. 2021, 10, 273. [Google Scholar] [CrossRef]
  10. Ningsih, D.F.; Junaid, A.; Mursalim, M. Environmental audit analysis to support Sustainability Development. Point View Res. Account. Audit. 2020, 1, 101–109. [Google Scholar] [CrossRef]
  11. Cai, C.; Zheng, K.; Wang, P. Research on the impact of government environmental audit on corporate environmental governance. Audit. Res. 2021, 4, 3–13. [Google Scholar]
  12. Kyzdarbekova, A.; Nurmagambetova, A.; Nurgaliyeva, A.; Bimendiyeva, L.; Assanova, A.; Mirzhakypova, S. Evaluating the role of environmental auditing in green economic growth in Kazakhstan. Herit. Sustain. Dev. 2025, 7, 179–196. [Google Scholar] [CrossRef]
  13. Huang, R.; Xie, X. Outgoing audit of natural resource assets of leading cadres and regional environmental quality: From the perspective of natural resource endowment. Resour. Sci. 2023, 45, 1091–1106. [Google Scholar]
  14. Guan, Y.; Xing, X. The Impact of Government Environmental Audit on the Quality of Corporate Environmental Information Disclosure. Acad. J. Bus. Manag. 2023, 5, 182–189. [Google Scholar]
  15. Zeng, C.; Zhang, L.; Li, J. The impact of top management’s environmental responsibility audit on corporate environmental investment: Evidence from China. Sustain. Account. Manag. Policy J. 2020, 11, 1271–1291. [Google Scholar] [CrossRef]
  16. Zhang, W.L.; Wang, X.M. Government audit, innovative activities, and green development. Technol. Anal. Strateg. Manag. 2025, 37, 298–311. [Google Scholar] [CrossRef]
  17. Jiang, Q.; Tan, Q. Can government environmental auditing improve static and dynamic ecological efficiency in China? Environ. Sci. Pollut. Res. 2020, 27, 21733–21746. [Google Scholar] [CrossRef] [PubMed]
  18. Cook, W.; van Bommel, S.; Turnhout, E. Inside environmental auditing: Effectiveness, objectivity, and transparency. Curr. Opin. Environ. Sustain. 2016, 18, 33–39. [Google Scholar] [CrossRef]
  19. Chaturvedi, S.; Parmar, S. The Green Economy and Sustainability: How Do Climate Change Policies Affect Indian Stock Market Returns? In Adaptive Strategies for Green Economy and Sustainability Policies; IGI Global Scientific Publishing: Hershey, PA, USA, 2025; pp. 405–424. [Google Scholar]
  20. Zhou, P.; Ang, B.W.; Poh, K.L. A survey of data envelopment analysis in energy and environmental studies. Eur. J. Oper. Res. 2008, 189, 1–18. [Google Scholar] [CrossRef]
  21. Song, M.; Fisher, R.; Kwoh, Y. Technological challenges of green innovation and sustainable resource management with large scale data. Technol. Forecast. Soc. Change 2019, 144, 361–368. [Google Scholar] [CrossRef]
  22. Chen, W.; Yao, L. Evaluating the carbon total factor productivity of China: Based on Cobb–Douglas production function. Environ. Sci. Pollut. Res. 2024, 31, 13722–13738. [Google Scholar] [CrossRef]
  23. Stjepanović, S.; Tomić, D.; Škare, M. Policy lessons from green gdp convergence over five decades: Enhancing sustainability and economic outcomes. J. Policy Model. 2024, 47, 187–210. [Google Scholar] [CrossRef]
  24. Wang, Y.; Sun, X.; Guo, X. Environmental regulation and green productivity growth: Empirical evidence on the Porter Hypothesis from OECD industrial sectors. Energy Policy 2019, 132, 611–619. [Google Scholar] [CrossRef]
  25. Chen, W.; Zou, W.; Zhong, K.; Aliyeva, A. Machine learning assessment under the development of green technology innovation: A perspective of energy transition. Renew. Energy 2023, 214, 65–73. [Google Scholar] [CrossRef]
  26. Munitlak-Ivanovic, O.; Zubovic, J.; Mitic, P. Relationship between sustainable development and green economy: Emphasis on green finance and banking. Èkon. Poljopr. 2017, 64, 1467–1482. [Google Scholar] [CrossRef]
  27. Kaseng, E.S. Kampanye Digital dalam Meningkatkan Kesadaran Masyarakat terhadap Isu Lingkungan: Digital Campaign to Increase Public Awareness of Environmental Issues. J. Marg. Soc. Res. 2024, 1, 68–72. [Google Scholar]
  28. Khan, M.A.; Ghouri, A.M. Environmental pollution: Its effects on life and its remedies. Res. World J. Arts Sci. Commer. 2011, 2, 276–285. [Google Scholar]
  29. Gunningham, N. Environmental auditing: Who audits the auditors? Environ. Plan. Law J. 1993, 10, 229. [Google Scholar]
  30. Lewis, L. Environmental audits in local government: A useful means to progress in sustainable development. Account. Forum. 2000, 24, 296–318. [Google Scholar] [CrossRef]
  31. Jiang, C.; Liu, R.; Han, J. Does accountability audit of natural resource promote corporate environmental performance? An external supervision perspective. Environ. Dev. Sustain. 2023, 25, 9417–9438. [Google Scholar] [CrossRef]
  32. Ye, C. Financial and Accounting Supervision and Audit Supervision: From the Perspective of Entrusted Economic Responsibility; Chinese Certified Public Accountant: Beijing, China, 2021; pp. 75–77. [Google Scholar]
  33. Liu, J.Y. National Audit in the process of National Governance Modernization: Institutional guarantee and practical logic. Chin. Soc. Sci. 2015, 9, 64–83. [Google Scholar]
  34. Wan, L.; Zheng, H.; Kong, D. Methodological innovation in government environmental auditing through biomechanical principles: An approach to environmental impact performance evaluation. Mol. Cell. Biomech. 2025, 22, 1704. [Google Scholar] [CrossRef]
  35. Damiyano, D. Sustainable Social Services Financing for Local Governments. In Financial Sustainability of Local Governments in Southern Africa; Routledge: Abingdon, UK, 2025; pp. 198–214. [Google Scholar]
  36. Rabasco, M.; Battiston, P. Predicting the deterrence effect of tax audits. A machine learning approach. Metroeconomica 2023, 74, 531–556. [Google Scholar] [CrossRef]
  37. Huang, H.; Yang, J.; Ren, C. The Impact and Mechanisms of Artificial Intelligence on Green Economic Efficiency: Empirical Evidence from China’s GTFP Improvement. J. Knowl. Econ. 2025, 2025, 1–35. [Google Scholar] [CrossRef]
  38. Lu, D.; Wang, Z. Towards green economic recovery: How to improve green total factor productivity. Econ. Change Restruct. 2023, 56, 3163–3185. [Google Scholar] [CrossRef]
  39. Bigsten, A.; Kayizzi-Mugerwa, S. Adaptation and distress in the urban economy: A study of Kampala households. World Dev. 1992, 20, 1423–1441. [Google Scholar] [CrossRef]
  40. Johnston, G.; Crombie, I.K.; Alder, E.M.; Davies, H.T.O.; Millard, A. Reviewing audit: Barriers and facilitating factors for effective clinical audit. BMJ Qual. Saf. 2000, 9, 23–36. [Google Scholar] [CrossRef] [PubMed]
  41. Zhang, Y.; Zhang, R.G.; Sun, S. Research on the impact of resource and environmental audit on high-quality economic development—From the perspective of national governance. Econ. Issues 2022, 106–114. Available online: https://guoyanwang.clcn.net.cn/DRCNet.Mirror.Documents.Web/DocSummary.aspx?DocID=6712014&leafID=14100 (accessed on 25 March 2025).
  42. Becker, S.O.; Ichino, A. Estimation of Average Treatment Effects Based on Propensity Scores. Stata J. 2002, 2, 358–377. [Google Scholar] [CrossRef]
  43. Liu, Y.; Zhang, X.; Lan, X. The impact of mixed ownership reform of state-owned enterprises on total factor productivity: An empirical study based on PSM-DID method. Financ. Res. 2016, 10, 63–75. [Google Scholar] [CrossRef]
  44. Roth, J. Pretest with Caution: Event-Study Estimates after Testing for Parallel Trends. Am. Econ. Rev. Insights 2022, 4, 305–322. [Google Scholar] [CrossRef]
  45. Peisakhin, L.; Rozenas, A. Electoral Effects of Biased Media: Russian Television in Ukraine. Am. J. Political Sci. 2018, 62, 535–550. [Google Scholar] [CrossRef]
  46. Chen, X.; Zhu, L.; Wang, Y. Resident effect: Empirical evidence from national land supervision. Economics 2019, 18, 99–122. [Google Scholar]
  47. Zhang, T.; Li, J. Network infrastructure, inclusive green growth and regional disparity: Causal inference based on dual machine learning. Quant. Econ. Tech. Econ. Res. 2023, 40, 113–135. [Google Scholar]
  48. Wang, B.; Liu, G. Energy conservation and emission reduction and China’s green economic growth—From the perspective of total factor productivity. China Ind. Econ. 2015, 5, 57–69. [Google Scholar]
  49. Jiang, T. Mediation and moderation effects in empirical research on causal inference. China Ind. Econ. 2015, 5, 100–120. [Google Scholar]
  50. Pei, M.; Liu, N. Can national urban agglomeration planning promote urban green innovation? Mod. Econ. Res. 2024, 6, 16–26. [Google Scholar]
  51. Dong, X.; Song, S.; Zhu, H. Industrial structure and economic fluctuation—Evidence from China. Soc. Sci. J. 2011, 48, 468–477. [Google Scholar] [CrossRef]
  52. Jia, J.; Guo, Q.; Zhang, J. Fiscal decentralization and local expenditure policy in China. China Econ. Rev. 2014, 28, 107–122. [Google Scholar] [CrossRef]
  53. Guo, P.; Wang, G. The synergistic effect and mechanism of digital economy in reducing pollution and carbon emissions: An empirical test based on prefecture-level city data. Resour. Sci. 2023, 45, 2117–2129. [Google Scholar]
  54. Wu, X.; Cao, Q.; Tan, X.; Li, L. The effect of audit of outgoing leading officials’ natural resource accountability on environmental governance: Evidence from China. Manag. Audit. J. 2020, 35, 1213–1241. [Google Scholar] [CrossRef]
  55. Zhang, H.; Lai, J.; Kang, C. Green signalling under environmental pressure: Does local government environmental regulatory pressure promote corporate environmental information disclosure? Econ. Anal. Policy 2024, 83, 813–844. [Google Scholar] [CrossRef]
  56. Xu, Y.; Tang, M.; Wang, M. Research on the impact of government environmental audit on green total factor productivity. Account. Mon. 2024, 45, 86–92. [Google Scholar]
  57. Porter, M.E.; Linde, C. Toward a new conception of the environment-competitiveness relationship. J. Econ. Perspect. 1995, 9, 97–118. [Google Scholar] [CrossRef]
Figure 1. Research framework and workflow of the impact of government environmental audits on urban green economy.
Figure 1. Research framework and workflow of the impact of government environmental audits on urban green economy.
Sustainability 17 05289 g001
Figure 2. The dynamic impact of government environmental audits on urban green economic.
Figure 2. The dynamic impact of government environmental audits on urban green economic.
Sustainability 17 05289 g002
Figure 3. Placebo test: A placebo test of the impact of government environmental audits on urban green economy.
Figure 3. Placebo test: A placebo test of the impact of government environmental audits on urban green economy.
Sustainability 17 05289 g003
Figure 4. The mechanism of government environmental audit influencing urban green economy.
Figure 4. The mechanism of government environmental audit influencing urban green economy.
Sustainability 17 05289 g004
Table 1. Hypothesis derivation table.
Table 1. Hypothesis derivation table.
Hypothesis NumberCore Content
H1Government environmental audit significantly improves the performance of urban green economy
H2Government environmental audit promotes green economic performance by promoting technological progress by promoting green innovation
H3Government environmental audit can promote the development of green economy by optimizing industrial structure and improving efficiency
H4Fiscal decentralization positively regulates the impact of government environmental audits on green economy
Table 2. Main variable names and calculation methods.
Table 2. Main variable names and calculation methods.
Variable CategoryVariable NameVariable SymbolCalculation MethodData Source
Dependent variableGreen economyGtfpMeasured using the SBM model—GML indexChina Urban Statistical Yearbook, Energy Statistical Yearbook, Environmental Statistical Yearbook
Key independent variablePerformance audit of key energy conservation and environmental protection fundsTreati × PosttDummy variable (0, 1)Official announcement by the National Audit Office of China (2017 audit results)
Control variablesEconomic development levelEcoLogarithm of city GDPChina Urban Statistical Yearbook
Industrial structure levelConRatio of tertiary industry-added value to secondary industry-added value in each cityChina Urban Statistical Yearbook
Government intervention levelFisRatio of general budgetary fiscal expenditure to GDP in each cityChina Urban Statistical Yearbook
Digitalization levelDigPeking University Digital Financial Inclusion Index divided by 100Peking University Digital Finance Research Center
Financial support levelSizeRatio of year-end financial loan balance to GDP in each cityChina Urban Statistical Yearbook
Mediating variablesGreen innovationLscxNumber of green innovation patents granted per 10,000 peopleChina National Intellectual Property Administration (CNIPA), processed by the authors
Industrial structure rationalizationTheiMeasured using the Theil IndexDerived from industry output data in China Urban Statistical Yearbook
Moderating variableFiscal decentralizationCzfqRatio of general fiscal revenue to general fiscal expenditure in each cityChina Urban Statistical Yearbook
Table 3. Variance inflation factor test.
Table 3. Variance inflation factor test.
VariableVIF1/VIFVariableVIF1/VIF
Treati × Postt1.400.7137Fis2.500.3996
Eco2.440.4104Dig1.450.6893
Con1.820.5498Size1.500.6653
Table 4. Propensity score matching results.
Table 4. Propensity score matching results.
VariableSampleMeanStandardized Bias (%)Reduction in Bias (%)
Treatment GroupControl Group
EcoBefore matching12.6612.5323.687.6
After matching12.6612.67−2.90
ConBefore matching1.310.9760.789.8
After matching1.311.35−6.20
FisBefore matching0.220.2022.694.5
After matching0.220.221.30
DigBefore matching2.031.4999.396.2
After matching2.032.003.80
SizeBefore matching1.260.9943.090.8
After matching1.261.28−4.0
Table 5. Descriptive statistics results.
Table 5. Descriptive statistics results.
VariableObservationsMeanStandard DeviationMinimumMaximum
Gtfp24731.00760.05010.37341.5075
Treati × Postt24730.37440.484101
Eco247312.58090.536310.745614.2358
Con24731.10060.56500.18885.3482
Fis24730.20440.10200.04390.9155
Dig24731.69190.60720.18014.0663
Size24731.09330.61690.16797.4502
Table 6. Baseline regression results.
Table 6. Baseline regression results.
Variable(1)(2)(3)
Green EconomyGreen EconomyGreen Economy
Treati × Postt0.0120 ***0.0125 ***0.0112 ***
(0.0042)(0.0042)(0.0042)
Eco −0.0129−0.0126
(0.0081)(0.0083)
Con −0.0109 **−0.0109 **
(0.0045)(0.0045)
Fis −0.0534−0.0601 *
(0.0360)(0.0362)
Dig 0.01584 ***
(0.0057)
Size 0.0001
(0.0041)
Constant1.0031 ***1.1887 ***1.1596 ***
(0.0018)(0.1067)(0.1094)
Observations247324732473
R-squared0.2450.2490.251
City FEYESYESYES
Year FEYESYESYES
Note: *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively. The standard errors are reported in parentheses. The same applies hereafter.
Table 7. Robustness test: outlier test, replacement of explained variables.
Table 7. Robustness test: outlier test, replacement of explained variables.
Variable(1)(2)
Green EconomyGreen Economy
Treati × Postt0.0101 ***0.0045 **
(0.0033)(0.0018)
ControlsYESYES
Constant1.1326 ***1.0593 ***
(0.0907)(0.0463)
Observations24732473
R-squared0.2900.244
City FEYESYES
Year FEYESYES
Note: ** and *** indicate significance at the 5% and 1% levels, respectively, with standard errors in parentheses. The same applies hereafter.
Table 8. Robustness test: Changing the sample period, double machine learning replacement model.
Table 8. Robustness test: Changing the sample period, double machine learning replacement model.
(1)(2)(3)
VariableGreen EconomyGreen EconomyGreen Economy
Treati × Postt0.0126 ***0.0149 ***0.0149 ***
(0.0048)(0.0048)(0.0029)
ControlsNOYESYES
Constant1.0085 ***0.5396 ***−0.0117 ***
(0.0020)(0.1937)(0.0012)
Observations159815982473
R-squared0.2640.277
City FEYESYESYES
Year FEYESYESYES
*** indicate significance at the 1% levels, respectively, with standard errors in parentheses.
Table 9. Robustness test: Excluding the impact of green finance reform pilot innovation zone policies.
Table 9. Robustness test: Excluding the impact of green finance reform pilot innovation zone policies.
(1)(2)
VariableGreen EconomyGreen Economy
Treati × Postt0.0140 ***0.0137 ***
(0.0042)(0.0043)
ControlsNOYES
Constant1.0032 ***1.1666 ***
(0.0018)(0.1185)
Observations21822182
R-squared0.2490.255
City FEYESYES
Year FEYESYES
*** indicate significance at the 1% levels, respectively, with standard errors in parentheses.
Table 10. Mechanism testing: green innovation and industrial structure optimization.
Table 10. Mechanism testing: green innovation and industrial structure optimization.
(1)(2)
VariableGreen InnovationRationalization of Industrial Structure
Treati × Postt0.1750 ***0.0250 ***
(0.0586)(0.0082)
ControlsYESYES
Constant11.1276 ***0.8820 ***
(1.5081)(0.2113)
Observations24732473
R-squared0.8060.825
City FEYESYES
Year FEYESYES
*** indicate significance at the 1% levels, respectively, with standard errors in parentheses.
Table 11. Moderating effect: fiscal decentralization.
Table 11. Moderating effect: fiscal decentralization.
(1)(2)
VariableGreen EconomyGreen Economy
Treati × Postt0.0121 ***−0.0044
(0.0043)(0.0062)
Czfq−0.0320 *−0.0354 *
(0.0190)(0.0190)
Czfq × Treati × Postt 0.0388 ***
(0.0106)
ControlsYESYES
Constant1.1480 ***1.0609 ***
(0.1096)(0.1118)
Observations24732473
R-squared0.2520.257
City FEYESYES
Year FEYESYES
Note: * and *** indicate significance at the 10% and 1% levels, respectively, with standard errors in parentheses. The same applies hereafter.
Table 12. Heterogeneity analysis: fiscal transparency, environmental regulation.
Table 12. Heterogeneity analysis: fiscal transparency, environmental regulation.
(1)(2)(3)(4)
VariableLow Level of Fiscal TransparencyHigh Level of Fiscal TransparencyLow Level of Environmental RegulationHigh Level of Environmental Regulation
Treati × Postt0.0176 ***0.00730.00840.0169 ***
(0.0064)(0.0085)(0.0072)(0.0064)
ControlsYESYESYESYES
Constant1.0663 ***0.8634 ***1.0055 ***1.1267 ***
(0.1536)(0.2888)(0.1563)(0.1995)
Observations1129113312151218
R-squared0.3260.3250.2750.337
City FEYESYESYESYES
Year FEYESYESYESYES
*** indicate significance at the 1% levels, respectively, with standard errors in parentheses.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Li, X.; Dong, B.; Li, S.; Xie, B.; Xie, L. Accelerating Green Growth: The Impact of Government Environmental Audits on Urban Green Economy. Sustainability 2025, 17, 5289. https://doi.org/10.3390/su17125289

AMA Style

Li X, Dong B, Li S, Xie B, Xie L. Accelerating Green Growth: The Impact of Government Environmental Audits on Urban Green Economy. Sustainability. 2025; 17(12):5289. https://doi.org/10.3390/su17125289

Chicago/Turabian Style

Li, Xinyu, Bingrui Dong, Shujuan Li, Bangsheng Xie, and Luhua Xie. 2025. "Accelerating Green Growth: The Impact of Government Environmental Audits on Urban Green Economy" Sustainability 17, no. 12: 5289. https://doi.org/10.3390/su17125289

APA Style

Li, X., Dong, B., Li, S., Xie, B., & Xie, L. (2025). Accelerating Green Growth: The Impact of Government Environmental Audits on Urban Green Economy. Sustainability, 17(12), 5289. https://doi.org/10.3390/su17125289

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