Sustainability Effects of Free Trade Zones: Evidence from Water Pollution in China

Abstract

Under the collaborative framework of sustainable development and environmental pollution control in China, there is an urgent need to break the governance dilemma of traditional environmental regulations and explore innovative paths for sustainability. This paper empirically tests the direct impact, spatial spillover effects, and mechanisms of free trade zones (FTZs) in China in reducing water pollution. Using a spatial Durbin model (SDM) combined with the staggered difference-in-differences (STA-DID) method on a dataset of 266 Chinese cities encompassing eastern, central, and western regions with diverse economic and environmental baselines from 2003 to 2023, the study finds that FTZs significantly reduce local water pollution by 9.17 million tons of untreated sewage discharge (β = −916.6, p < 0.01), with a spatial spillover effect that decreases pollution in surrounding cities by 12.33 million tons (β = −1232.9, p < 0.01). Upgrading industrial structure, accelerating technological innovation, and strengthening government environmental governance constitute the core mediating channels. This study provides theoretical support for institutional innovation in environmental governance and empirical evidence to address the trade-off between economic growth and environmental protection in China, contributing to the understanding of how context-specific institutional innovations can advance regional sustainability, aligning with the United Nations Sustainable Development Goals (SDGs).

1. Introduction

Water pollution, a key factor constraining global sustainable development, has transcended a single ecological issue to evolve into a complex crisis threatening human health, disrupting ecological balance, and restricting economic growth [1,2]. A World Health Organization (WHO) report states that 80% of global diseases are associated with contaminated water. In China, over 90% of urban water bodies are polluted, with 170 million people drinking water contaminated by organic matter and 300 million urban residents facing water pollution issues. In 2022, 10.6% of surface water monitoring sections in China failed to meet drinking water standards, and the exceedance rate of heavy metals in groundwater in North China reached 18.3% (China National Environmental Monitoring Centre, 2022). These circumstances highlight that water pollution has become a critical bottleneck, restricting China’s achievement of Sustainable Development Goals, not only shaking the health foundation and ecological security of contemporary society, but also eroding the capacity of future generations to pursue a better life through intergenerational resource allocation imbalances.

In 2024, China’s investment in water pollution prevention and control reached CNY 26.7 billion, yet groundwater contamination remains poorly mitigated. Traditional environmental regulations, relying on administrative mandates and end-of-pipe treatment, have significantly increased corporate compliance costs, trapping water pollution governance in an “input–efficiency” imbalance. There is an urgent need for innovative governance models beyond conventional regulatory approaches. China’s free trade zones (FTZs) offer a unique institutional context to address this crisis. As policy pilots integrating economic openness with green governance, FTZs present an opportunity to test how regional institutional innovation balances development and environmental protection.

This raises the following critical questions: As a regional innovation policy driving economic development [3,4,5], can the FTZ improve water pollution in China and thereby promote coordination between economic development and environmental protection? Regarding this issue, we propose Hypothesis 1: the FTZ strategy can reduce the level of water pollution. Secondly, if there is an improvement effect, what is the specific mechanism? The upgrading of industrial structure, technological innovation, and government governance have been proven to effectively improve pollution [6,7,8], and these three factors are significantly associated with FTZs [9,10]. Therefore, regarding this issue, we propose Hypothesis 2: the FTZ strategy can reduce water pollution through facilitating industrial structure upgrading, fostering technological innovation and enhancing government environmental governance. Finally, regional policies often have spillover effects [11]. Do FTZs also have effective spatial radiation effects on improving water pollution? Regarding this issue, we propose Hypothesis 3: the policy effect of the FTZ strategy in reducing water pollution has positive spatial spillover effects.

To overcome the limitation of traditional methods ignoring spatial dependence [12], this paper uses the spatial Durbin model (SDM) combined with the staggered difference-in-differences (STA-DID) method to analyze a dataset of 266 cities in eastern, central, and western China from 2003 to 2023, in order to clarify the direct impact, mediating mechanism, and spatial characteristics of FTZs on water pollution in China. The research findings can resolve the contradiction between China’s “economic growth and environmental constraints”, provide a scientific basis for policymakers to optimize regional development plans and improve green policy toolkits, and help China accelerate the realization of “Clean Water and Sanitation” (SDG6) and “Sustainable Cities and Communities” (SDG11) in the Sustainable Development Goals (SDGs).

The remaining sections of this paper are arranged as follows: Section 2 is a literature review, summarizing existing research achievements, identifying the shortcomings of existing studies and clarifying the research focus and direction of this paper. Section 3 is the research hypothesis, analyzing the theoretical mechanism of the impact of FTZs on water pollution. Section 4 is the research design, constructing diversified empirical models. Section 5 is the empirical results and analysis, verifying relevant research hypotheses from the data level. Section 6 is the discussion. Section 7 is the research conclusion.

2. Literature Review

The above introduction has clearly defined the practical background and core objectives of this study. However, in order to achieve this goal and deepen the understanding of the research question, it is crucial to systematically review and analyze existing research results. Therefore, the following will comprehensively analyze the research progress that scholars have made in related fields through a literature review, summarize the research gaps and shortcomings, and provide a solid theoretical basis and research direction guidance for subsequent research content and design.

2.1. Research on Water Pollution

Firstly, in the area of water pollution control, existing studies indicate that control measures tailored to the natural geographical conditions of water resources in China have yielded relatively significant results. Specifically, in the context of river water pollution control, the “River Chief System” has played a crucial role in reducing key pollution indicators [13,14]. The “River Chief System” emphasizes a comprehensive governance approach, highlighting the integration and coordination of the water resource system with the human social system [15]. In the case of lake water pollution control, the “Lake Chief System” has notably enhanced the overall water quality in pilot lakes, significantly reducing the total amount of pollutants exceeding the standard [16]. In the governance of watershed water pollution, the watershed ecological compensation policy contributes to improving the level of green development in the watershed, promoting the rational utilization of water resources, reducing the intensity of water resource use, and thus enhancing water quality [17,18].

Secondly, in terms of the government’s role and function, the existing literature emphasizes the critical role played by government agencies in water pollution control. The establishment of water ecological civilization cities has significantly improved water quality, with COD and NH₃ as primary indicators [19,20]. The allocation of total water pollution emissions, managed by provincial and municipal governments, should balance fairness with efficiency to maximize the overall benefits of regional pollution control [21]. Moreover, the flattening of environmental protection agencies has helped curb local protectionism and official corruption, while enhancing the independence of these agencies and facilitating reductions in border pollution [22]. The implementation of audit pilots has also significantly improved water quality [23,24]. Finally, in terms of the relationship between water pollution control and social–economic development, as ecological civilization transformation and green development concepts are more deeply promoted, the relationship has gradually shifted from a “zero-sum” conflict to a “win-win” state of coordination [25,26]. It has also been shown that water pollution control can proceed concurrently with the stable development of the banking industry [27].

2.2. Research on the Policy Effects of FTZs

The existing literature generally agrees that FTZs play a positive role in promoting economic development. Several macro-level studies have analyzed the effects of FTZs on the economy and trade, highlighting that FTZs can effectively attract foreign investment [28,29], promote regional synergy and complementarity [30], align with high-standard international economic and trade rules, and foster economic growth [3]. The development of FTZs has been found to significantly contribute to overall economic growth [4], drive the upgrading and rationalization of industrial structures 9, generate regional spillover effects that stimulate local economic growth [31,32], enhance regional innovation [33,34], and improve the export resilience of cities [35]. In addition to promoting economic growth, FTZs are shown to facilitate trade development [36], reduce trade costs, and optimize resource allocation [37], thereby improving the structure of trade 5.

At the micro level, some studies have observed that FTZs positively impact firms by improving productivity [38,39], increasing export value added [40], enhancing trade facilitation [41], elevating the quality of economic development [5], and strengthening the resilience of industrial and supply chains [42]. FTZs have also been credited with improving the efficiency of financial services supporting the real economy [43], enhancing trade efficiency [32], and promoting the expansion of non-primary industries [44].

Very few studies have explored the relationship between FTZ and environmental pollution and have only focused on atmospheric pollution. Research has found that the establishment of China’s free trade zones can reduce air pollution concentrations in cities by 12–17% [45].

2.3. Summary

Through an inductive analysis of the existing literature, it becomes evident that current research on water pollution primarily focuses on the effectiveness of water pollution control policies and the role of government in the governance process. While achieving coordinated development between water pollution control and social–economic growth has become an unquestionable future direction, there remain notable gaps in the literature. Most studies concentrate on macro-level environmental governance policies, while assessments of the environmental effects of other external economic policies are relatively scarce, particularly in the context of FTZ policies. Regarding research on FTZ strategies, the existing literature predominantly addresses their macroeconomic impacts or their effects on micro-firms. However, the existing literature lacks in-depth discussions on the environmental impact of FTZs and pays relatively little attention to their role in promoting regional sustainable development. Furthermore, most studies focus on the localized effects of FTZs, with limited exploration of their spatial spillover effects.

This paper systematically explores the intrinsic relationship between the FTZ strategy and water pollution control, which not only highly aligns with current academic trends, but also expands the cognitive boundaries of regional policy environmental effects, providing a new analytical dimension for solving the coordination problem between economic development and environmental protection. At the same time, the spatial Durbin model staggered difference-in-differences (SDM-STA-DID) method provides a scientific and practical technical paradigm for in-depth analysis of the environmental effects of policies.

3. Theoretical Mechanism and Research Hypothesis

This paper examines the impact of FTZs on water pollution levels, focusing on both the direct effects and spatial spillover effects, as well as the underlying mechanisms. Therefore, the theoretical mechanism and research hypothesis of this paper are as follows.

3.1. Direct Impact

The preceding discussion illustrates that the FTZs can effectively reduce water pollution through three key aspects. Firstly, the green development framework of FTZs addresses water pollution at its source. Through comprehensive and in-depth environmental impact assessments, the overall industrial layout of FTZs is ensured to meet environmental protection requirements, avoiding water pollution problems caused by unreasonable project introductions and providing strong guarantees for the sustainable utilization of regional water resources and the protection of water environments. At the same time, stringent market access has avoided the entry of high-polluting and high-energy-consuming enterprises, as well as the clustering of traditional heavy polluting industries. Secondly, institutional innovation in FTZs refers to the systematic changes and creative reconstruction of institutional rules, management modes, and operational mechanisms in the areas of trade, investment, finance, and regulation within FTZs, with the goal of breaking through traditional institutional constraints and constructing a new economic system at a higher level. Its core lies in breaking policy barriers through “system-oriented opening”, forming replicable and scalable reform experience and serving the national strategic objectives. China’s free trade zones have considerable autonomy in reform in accordance with the strategic needs of the country’s sustainable development. In the field of water pollution control, FTZs enhance corporate responsibility in water pollution control through institutional innovation. By utilizing market-driven mechanisms, such as the establishment of water rights trading markets, FTZs incentivize firms to reduce pollution emissions, thereby achieving a balance between environmental protection and economic growth. Finally, green finance (specifically including green credit, green bonds, green funds, etc.) refers to activities that support environmentally sustainable development through financial means. In China, FTZs, as an important platform for opening up to the outside world, have assumed the mission of promoting the development of green finance. Therefore, FTZs provide sufficient financial support for water pollution control by promoting the development of green finance. The financing guarantee system established by green finance projects such as green bonds and green funds effectively ensures the smooth implementation of various water pollution control works.

At the same time, from a global perspective, with the continuous improvement of global environmental standards, FTZs must comply with the higher requirements of the international market for environmental protection to avoid international trade frictions, promote enterprises in the zone to adapt to stricter environmental standards, and thereby reduce pollution of water resources.

FTZs are committed to achieving the harmonious integration of ecological, economic, and social benefits in the development process, promoting the gradual shift in socioeconomic development from the past “zero-sum” confrontation to “win-win” outcomes [45]. With the deepening of ecological civilization initiatives and the full incorporation of green development principles, FTZs are expected to play an even more active role in protecting water resources and improving water quality, providing strong support for achieving the long-term goal of sustainable development.

On the basis of the foregoing analysis, this paper puts forward the following hypothesis:

Hypothesis 1. 

The FTZ strategy can reduce the level of water pollution.

3.2. Indirect Mechanism

3.2.1. FTZs Reduce Water Pollution by Facilitating Industrial Structure Upgrading

As an important practice in trade liberalization, the establishment of FTZs has accelerated the upgrading of industrial structures [9]. Firstly, through technology diffusion and knowledge transfer, incoming high-tech firms catalyze the industrial upgrading of local enterprises, driving their evolution from labor-intensive to technology-intensive production paradigms [46]. Secondly, the presence of foreign high-tech firms not only increase local competitiveness but also exerts a “crowding-out effect” on pollution-intensive industries [47]. Finally, by revising the criteria for evaluating trade advantages, FTZs help prevent the cross-regional relocation of hidden pollution, enhance the internalization of environmental costs, and rectify the discrepancies between firm decision-making and social efficiency. This mechanism effectively reduces the share of high-energy-consuming industries and promotes the continued upgrading of the industrial structure [48].

The upgrading of the industrial structure is a key factor in promoting the turning of the Environmental Kuznets Curve [6]. By improving the environmental protection technology level of the industry, reducing water resource demand, and enhancing water resource utilization efficiency, the upgrading of the industrial structure provides important support for improving water quality and reducing water pollution. Regarding the improvement of protection technology in industries, the production process of traditional industry is often accompanied by a large amount of wastewater discharge, which may contain harmful substances such as heavy metals and chemical agents, causing serious pollution to the water environment. With the adjustment of the industrial structure, the service industry, represented by information technology and financial services, and the high-tech industry, centered on researching new energy and new materials, are flourishing. The amount and pollutant composition of wastewater generated during the production and operation of these industries are far lower than those of traditional industries [49]. Green industries, such as ecological agriculture, widely adopt efficient water resource utilization technologies and advanced wastewater treatment technologies to curb water pollution from the source and lay a solid foundation for improving water quality [50]. Regarding reducing water demand and improving water use efficiency, the development of new industries has promoted change in the demand pattern for water resources. Traditional industries often require a large amount of water resources for production cooling and cleaning, and the reuse rate of water resources is relatively low [51]. The emerging high-tech industries and service sectors have significantly reduced their dependence on natural resources, while placing greater emphasis on efficient and circular utilization of resources [52]. These industries, with knowledge and information at their core, hardly involve large-scale water resource consumption in their operation processes, reducing the overall demand pressure on water resources and indirectly reducing the risk of water pollution that may be caused by the excessive development and use of water resources.

On the basis of the foregoing analysis, this paper puts forward the following hypothesis:

Hypothesis 2.1. 

The FTZ strategy can reduce water pollution through facilitating industrial structure upgrading.

3.2.2. FTZs Reduce Water Pollution by Fostering Technological Innovation

FTZs have been shown to significantly foster technological innovation. Firstly, the FTZ strategy has accelerated the marketization process, which in turn has strengthened the incentive effect of firm R&D investment by promoting “capacity reduction” and green upgrading of industrial production methods, resulting in a substantial synergistic effect between market dynamics and research development [53]. Moreover, institutional innovations within FTZs have provided strong support for technological advancements. For instance, the emissions trading scheme implemented in FTZs has effectively encouraged firms to innovate in the area of green invention patents [54]. Additionally, FTZs have further advanced green technology by optimizing foreign investment policies and attracting cutting-edge foreign technologies [55].

Technological innovation, particularly advancements in water treatment, clean production, and water-saving technologies, has proven effective in reducing water pollution in numerous cases. On one hand, the widespread adoption of clean production processes and water-saving technologies has significantly reduced wastewater discharge during production, thereby addressing water pollution at its source [7]. On the other hand, the promotion and implementation of advanced water treatment technologies have effectively controlled the discharge of both industrial and domestic sewage, substantially alleviating the pollution burden [56,57].

On the basis of the foregoing analysis, this paper puts forward the following hypothesis:

Hypothesis 2.2. 

The FTZ strategy can reduce water pollution through fostering technological innovation.

3.2.3. FTZs Reduce Water Pollution by Enhancing Government Environmental Governance

Through a series of reform measures and policy innovations, FTZs have played a crucial role in environmental governance, offering a valuable opportunity to enhance the government’s environmental management capabilities. To attract foreign investment and meet the high environmental standards of the international market, FTZs have implemented stricter environmental regulations. This not only imposes constraints on local enterprises but also prompts local governments to increase their efforts in environmental governance, with the most effective means being extensive and proactive environmental investment. Research has shown that government investments in areas such as industrial wastewater treatment infrastructure and sewage network development can effectively reduce pollution discharge and improve water quality [58]. Additionally, the government promotes the research and application of clean technology and environmental governance technology through financial support, enabling firms to adopt pollution control measures that further reduce water pollution [8].

On the basis of the foregoing analysis, this paper puts forward the following hypothesis:

Hypothesis 2.3. 

The FTZ strategy can reduce water pollution through enhancing government environmental governance.

3.3. Spatial Effects

Firstly, the FTZs have formed a “demonstration effect” on neighboring non pilot cities through the spillover empowerment of industrial structure optimization and green technology innovation, FTZs attract the agglomeration of green energy, high-end manufacturing, and other enterprises through policy dividends, forming a “demonstration effect” on adjacent non-pilot cities. On the one hand, through supply chain linkages, industrial chain extension, and industrial transfer, FTZs promote surrounding cities to eliminate high-pollution and high-water-consuming industries, realizing the green transformation of industrial structures. For example, after the Tesla Gigafactory in the Shanghai FTZ went into operation, supporting links such as battery component shell manufacturing and interior parts production were gradually deployed in surrounding cities. Taking the opportunity of industrial undertaking, Suzhou, Jiaxing, and other cities guided traditional manufacturing enterprises to transform into green industries such as new energy vehicle components and high-end electronic materials, effectively reducing local sewage discharge. On the other hand, the green technology innovation achievements of FTZ enterprises, such as new water-saving processes and sewage treatment technologies, can be disseminated through technical cooperation and other means, driving surrounding cities to enhance their sewage treatment capabilities [45].

Secondly, there is a spillover effect of environmental regulations. Previous studies have confirmed that environmental regulations can not only reduce industrial pollution in pilot areas but also have spillover effects on surrounding areas through “competitive imitation effects” and “collaborative governance” [58,59,60]. The strict environmental regulation system within the FTZ, such as upgrading wastewater discharge standards and strengthening pollution monitoring measures, will encourage surrounding areas to actively increase regulatory intensity as regional integration progresses. For example, neighboring governments may proactively raise environmental standards to avoid a decline in industrial competitiveness. In addition, local enterprises will also require their peripheral enterprises in the supply chain to comply with environmental protection requirements and achieve unified regulatory standards through cross-regional environmental governance agreements, thereby forming a regional collaborative pollution control pattern and expanding the spatial coverage of water pollution control.

Thirdly, the learning effect occurs, as advanced environmental protection policies (emission trading, green financial instruments, ecological compensation mechanisms) and environmental technologies (membrane separation sewage treatment equipment, circulating water utilization systems) piloted in FTZs can be borrowed and implemented by surrounding areas through policy promotion, enterprise demonstration, technology transfer, and other means, thereby driving the improvement of water pollution in surrounding regions and achieving environmental governance enhancement in a broader scope. For instance, the “Water Pollution Joint Prevention and Control” mechanism of the Yangtze River Delta FTZs has been adopted by surrounding cities such as Suzhou and Wuxi, significantly improving the efficiency of regional water environment governance.

Fourthly, as a reform demonstration zone, FTZs can attract and gather a large number of development factors such as capital, technology, and talent, which will spread to surrounding areas through the flow and mismatch with the strengthening of regional economic integration. For example, when environmental protection enterprises establish production bases in FTZs, advanced pollution prevention technologies and management experience will spill over to surrounding areas, and the flow of high-end environmental protection talents can promote the improvement of environmental governance plans in surrounding cities. This flow of factors promotes the reconfiguration of regional resources, indirectly enhancing the effectiveness of water pollution control in surrounding areas.

On the basis of the foregoing analysis, this paper puts forward the following hypothesis:

Hypothesis 3. 

The policy effect of the FTZ strategy in reducing water pollution has positive spatial spillover effects.

To present the influence path more intuitively, the following influence framework diagram (Figure 1) is constructed based on the comprehensive analysis above:

4. Methodology

The above research hypotheses are logical deductions based on existing theories and the characteristics of the research question, laying a directional foundation for the study. But whether the hypothesis is valid and whether there is an inherent relationship between variables still needs to be verified through scientifically rigorous empirical analysis of the model. Therefore, appropriate research methods will be applied, appropriate data will be selected, corresponding models will be constructed, systematic testing of research hypotheses will be conducted, and persuasive conclusions will be drawn to further deepen the understanding of the research problem.

4.1. Variables

4.1.1. Dependent Variable

This paper uses the discharge of untreated sewage to represent the water pollution situation of each city (Water). This indicator refers to the volume of sewage directly discharged into natural water bodies within the city without effective treatment by sewage treatment facilities. The sources of pollution include wastewater generated by residents’ daily lives and industrial and commercial activity wastewater collected through municipal pipelines. The unit is 10,000 tons.

Compared to the indicators selected in previous studies, this indicator has the following advantages: Firstly, it can directly reflect the scale of the pollution source, accurately measure the degree of urban water pollution, and fundamentally reflect the pollution status of urban water resources. Secondly, it is sourced from official statistics. Not only is the statistical caliber clear and standardized, but the data acquisition process is relatively objective. Thirdly, this indicator, which has extensive representativeness, can comprehensively reflect the water pollution levels of different cities and is conducive to conducting comparative analysis and overall research.

4.1.2. Independent Variable

Using the dummy variable of the FTZ strategy ($FTZ$) as the independent variable, this variable is implemented by constructing the interaction term between the dummy variables of the implementation object (treat) and the implementation time (post). Object dummy variable (treat): The value of the city where the FTZ is established is 1, otherwise it is 0. Time dummy variable (post): The value before policy implementation is 0, and thereafter, it is 1.

4.1.3. Mediator Variables

Regarding the upgrading of industrial structure, this paper uses the ratio of the added value of the tertiary industry to the added value of the secondary industry for calculation (industry). This indicator can effectively reflect the trend of industrial structure transformation towards service-oriented and advanced industries, and its numerical changes intuitively reflect the dynamic shift in economic focus [61,62].

Regarding the innovation level, this paper measures it by the number of authorized patents (innovation). This index is not only comparable over time and across regions, but also can effectively reflect the innovation achievements and innovation development potential in various technical fields, which helps comprehensively demonstrate the innovation level [63,64].

Regarding government environmental governance, this paper measures it by government expenditure on environmental protection (governance). Environmental protection expenditure is an important indicator for measuring the government’s attitude and determination towards environmental governance. The scale of expenditure directly reflects the quantity of resources allocated by the government in environmental governance. At the same time, environmental protection expenditure is an effective indicator for determining whether governance policies have been implemented [65,66,67].

4.1.4. Control Variables

The degree of environmental pollution can be influenced by multiple factors such as the local economic development level, pollution control capacity, and population activities [68,69,70,71,72]. Based on relevant research, the following control variables are selected: (1) Urbanization rate (rate), expressed as the proportion of the urban population to the total population. (2) Population density (pop), measured by the number of residents per square kilometer. (3) Economic development level (pgdp), expressed in terms of per capita GDP. (4) Industrialization level (scale), expressed in terms of industrial output value. (5) Pollution control capacity (pipe), represented by the length of sewage pipelines.

4.2. Model

4.2.1. Basic Model

The general staggered difference-in-differences (STA-DID) method takes the following form:

(1)${Water}_{it}={\beta }_0+{\beta }_1{FTZ}_{it}+\lambda Z_{it}+{\nu }_i+{\phi }_t+{\epsilon }_{it}$

(2)${FTZ}_{it}={treat}_{it}\ast {post}_{it}$

In Equation (1), ${Water}_{it}$ is the level of water pollution. ${FTZ}_{it}$ is the estimator of DID. $Z_{it}$ is a series of control variables. ${\nu }_i$ and ${\phi }_t$ are the city and time effect, respectively. ${\epsilon }_{it}$ is the random disturbance term.

4.2.2. Spatial Effect Model

The spatial Durbin model (SDM) provides a multidimensional analytical perspective for analyzing spatial correlation characteristics by decomposing the total effect. In view of this, the spatial Durbin model–staggered difference-in-differences (SDM-STA-DID) method is constructed to accurately identify the local environmental governance effects of FTZs and quantitatively assess the spatial transmission intensity of policy innovations. The model takes the following form:

(3)${Water}_{it}={\beta }_0+{\beta }_1{FTZ}_{it}+\lambda Z_{it}+\rho {\sum }_{j=1}^N{W}_{ij}{Water}_{it}+\tau {\sum }_{j=1}^N{W}_{ij}{FTZ}_{it}+\gamma {\sum }_{j=1}^N{W}_{ij}{Z}_{it}+{\nu }_i+{\phi }_t+{\epsilon }_{it}$

(4)$W_{ij}=\left\{\begin{array}{c}{\displaystyle \frac{1}{d_{ij}}},i\ne j\\ 0,i=j\end{array}\right.$

In Equation (3), $\tau$ captures exogenous policy spillovers. In Equation (4), the matrix $W_{ij}$ is constructed based on the inverse geographical distance ($d_{ij}$) between each pair of cities. This 266 × 266 dimensional matrix captures the spatial dependence structure across all sample cities, with weights decreasing as inter-city distance increases, reflecting the distance decay effect in spatial interactions.

4.2.3. Influence Mechanism Model

Based on the mediating-effect testing paradigm [73], which has greater statistical superiority, this paper formulates the following mechanism-testing models, as specifically shown in Equations (5) and (6):

(5)${Med}_{it}={{\gamma }_1+{\gamma }_{2}FTZ}_{it}+{{\gamma }_{3}Z}_{it}+{\nu }_i+{\phi }_t+{\epsilon }_{it}$

(6)${Water}_{it}={{\tau }_0+{\tau }_{1}FTZ}_{it}+{{\tau }_{2}Med}_{it}{+{\tau }_{3}Z}_{it}+{\nu }_i+{\phi }_t+{\epsilon }_{it}$

In Equations (5) and (6), ${Med}_{it}$ represents a series of mediator variables. As indicated by the theoretical analysis presented earlier, the mediator variables consist of industrial structure upgrading (industry), technological innovation (innovation), and government environmental governance (governance).

4.3. Sample Selection and Data Sources

We select 266 prefecture-level cities as samples to ensure comprehensive geographical coverage and socioeconomic diversity across China. This selection accounts for over 90% of China’s economic output while maintaining a balanced representation of both FTZ and non-FTZ cities. This not only enables robust policy evaluation, ensuring that spatial econometric analyses can reliably detect spillover effects of FTZ policies, but also guarantees that research findings on the environmental governance effects of FTZs are applicable to China’s diverse urban contexts. The specific list of cities can be found in Appendix A. The time range is from 2003 to 2023. The year 2023 represents the most recent available data on water pollution, while 2013 marks the establishment of China’s first FTZs. This paper selects ten years before and after 2013 to ensure both the timeliness and representativeness of the data. The data in this paper comes from China Urban Statistical Yearbook and China Urban Construction Statistical Yearbook. Table 1 shows the descriptive statistics.

5. Empirical Results

5.1. Parallel Trend Test

Meeting the parallel trends assumption is a prerequisite for using the STA-DID method. Before the shock occurs, the change trends in water pollution across cities with and without FTZs should show a parallel tendency and have no systematic differences. If this condition is not met, the evaluation effect of the policy effect will be invalid. The model takes the following form:

(7)${Water}_{it}={\beta }_0+{\sum }_{t=-4}^{3+}{\beta }_k{{treat}_i\ast D}_{t_{io}+k}+\lambda Z_{it}+{\nu }_i+{\phi }_t+{\epsilon }_{it}$

In Equation (7), ${treat}_i$ indicates whether the city is an FTZ. $D_{t_{io}+k}$ represents a series of dummy variables. The variable $t_{io}$ represents the year when the FTZ strategy was implemented in city $i$. $k$ represents the difference between the year $t$ and the year $t_{io}$, dividing the sample time into three segments: before the implementation, during the implementation, and after the implementation. Taking the start year of the FTZ pilot as the base period, dummy variables for the four years prior to the base period and the three years following the base period are constructed to replace the original policy variable in Equation (1). According to the test results in Figure 2, the coefficients of each period before the current period crossed the 90% confidence interval indicated by the dashed line, indicating that samples from different groups maintained the same development trend before policy implementation.

5.2. Regression Results

5.2.1. Benchmark Regression

Table 2 shows the results of benchmark regression. The regression results of Equation (1) (Column 2) show that the estimated coefficient of $FTZ$ is statistically significant and negative (β = −916.6, p < 0.01). This empirical finding demonstrates that FTZ establishment has a significant inhibitory effect on local water pollution. Specifically, under the environmental optimization effect of the FTZ strategy, cities with FTZs demonstrate an annual reduction of approximately 9.16 million tons in untreated sewage discharge compared to cities without such strategies. This quantitative result provides strong empirical evidence supporting Hypothesis 1.

The regression results Equation (3) (Column 4) show that $W\ast FTZ$ is significantly negative, indicating that the impact of FTZs has externalities. This externality enables the improvement effect of FTZs on water pollution not only to be demonstrated within the pilot area but also to be extended to neighboring cities, thereby promoting a wider range of environmental improvement.

5.2.2. Decomposition of Spatial Effects

In the analytical framework of spatial econometric models, the impact of FTZs on water pollution requires breaking through the limitations of general regression models. General models often assume the independence of observational values across regions, whereas in reality, policy implementation in one region generates spillover effects on surrounding areas through spatial correlation. This interactive influence manifests as cross-partial derivative terms in the model, leading to deviations in interpreting policy effects directly through regression coefficients. Therefore, decomposing the total effect into direct and indirect effects via partial differentiation in spatial models becomes crucial for accurately revealing the policy impact mechanism [12,74].

From the logic of spatial models, the direct effect captures the net impact of FTZs on local water pollution. As shown in Table 3, the direct effect is −890.3, which means that the FTZ strategy reduces the annual discharge of untreated sewage in the local area by about 8.9 million tons. As previously analyzed, the generation of this local effect stems from the internal driving forces of FTZs, such as raising sewage discharge standards and promoting the green technological transformation of enterprises, which directly act on local pollution sources.

The indirect effect reveals the spatial spillover characteristics of the influence of FTZs. Within the spatial econometric framework, the strategic effects of FTZs are transmitted to neighboring regions through the spatial matrix, manifesting as multiple interactive effects in the model to form ripple-like diffusion impacts. As shown in Table 3, the indirect effect is -1232.9, indicating that after a city implements the FTZ strategy, the cumulative effect of spatial spillover can reduce the total untreated sewage discharge of all radiated regions by 12.329 million tons annually. As previously analyzed, whether due to the flow of production factors or pollution co-governance achieved through industrial transfer in economically complementary cities, the impacts of these processes are quantified as indirect effects through the iterative computation of spatial correlations in the model.

The core value of the spatial model lies in breaking the assumption of “regional independence” and treating impacts as a dynamic spatial interaction process. The distinction between direct and indirect effects is essentially a quantitative separation of the dual attributes of “local implementation” and “regional radiation” of FTZs. In this study, the significantly negative results of both effects indicate that the FTZs can not only directly reduce pollution through strict local governance but also drive collaborative pollution control in surrounding areas through spatial correlation mechanisms. The existence of this “dual effect” effectively confirms Hypothesis 3.

It is worth noting that the indirect effect of FTZs on improving water pollution is greater than the direct effect, and it dominates the total effect. This is consistent with the research results of many scholars; that is, the cumulative impact of spillover effects exceeds the direct impact of the local area [75,76]. Several potential explanations account for this observation: Firstly, FTZs have wide-ranging and continuously deep-going driving effects on surrounding cities. Surrounding cities have amplified the positive externalities of FTZs by undertaking the transfer of green industries and sharing environmental protection technologies. Secondly, the multiplier effect of technology diffusion is significant. After obtaining the technology, surrounding enterprises carry out secondary innovation and drive the upgrading of the industrial chain. Thirdly, the flow of factors stimulates the overall vitality of the region. Peripheral enterprises leverage the optimization of factors to strengthen water pollution control. The combined effect of multiple cities makes the scale and influence of the indirect effect exceed those of the direct effect.

As shown in Table 3, the total effect value rises to −2123.1 (direct effect + indirect effect), far higher than the estimated coefficient of the general model. Therefore, the general staggered difference-in-differences model underestimates the improvement capacity of FTZs on water pollution. From the perspective of practice, the effect decomposition of the spatial model provides a scientific basis for evaluating the global impact of policies, expanding the understanding of the environmental effects of FTZs from a single region to the entire spatial system and providing theoretical support for constructing a regional environmental governance system.

5.3. Robustness Test

5.3.1. Placebo Test

The robustness of our findings is further confirmed through a placebo test involving 500 random samplings. As illustrated in Figure 3, the kernel density distributions of both the estimated coefficients and p-values demonstrate three key characteristics: (1) The randomly generated coefficients follow a normal distribution centered around zero. (2) There is a significant gap between the actual estimated coefficients and the pseudo-effect clustering. (3) Most of the p-values exceed the 0.1 significance threshold. These patterns collectively validate that our core results are unlikely to be driven by spurious correlations or chance factors.

5.3.2. PSM-DID Test

The nonrandom placement of FTZs systematically favors cities with advanced infrastructure and greater economic openness, introducing potential selection bias in our estimation sample. This endogenous policy allocation mechanism, where treatment assignment (FTZ establishment) correlates with baseline city characteristics, may compromise the exogeneity assumption required for causal inference. To effectively address endogeneity problems, this study employs propensity score matching (PSM) for sample screening, ensuring that the different groups exhibit balanced characteristics.

Table 4 presents complementary evidence from PSM analyses, with Columns (2) and (4) reporting estimates using nearest-neighbor and caliper matching methods, respectively. The policy coefficients remain statistically significant (p < 0.01) and directionally consistent with our baseline specification, demonstrating that the water pollution reduction effect of FTZs persists across alternative matching algorithms. These robustness checks lend additional credibility to our core finding that FTZ implementation generates meaningful improvements in water quality.

5.3.3. Eliminate the RCS’s External Interference

In 2018, the Chinese government comprehensively promoted the strategy of the “River Chief System” (RCS). This river management system has effectively promoted water resource protection and water pollution prevention. Many studies have verified the significant improvement effect of the RCS on water pollution [77]. In order to separate the effect of the RCS on reducing water pollution, we adopt the following empirical framework:

(8)${Water}_{it}={\beta }_0+{\beta }_1{FTZ}_{it}+{\beta }_2{RCS}_{it}+\lambda Z_{it}+{\nu }_i+{\phi }_t+{\epsilon }_{it}$

(9)$$\begin{gathered} {Water}_{it}={\beta }_0+{\beta }_1{FTZ}_{it}+{\beta }_2{RCS}_{it}+\lambda Z_{it}+\rho {\sum }_{j=1}^N{W}_{ij}{water}_{it}+\tau {\sum }_{j=1}^N{W}_{ij}{FTZ}_{it}+\chi {\sum }_{j=1}^N{W}_{ij}{RCS}_{it}+ \\ \gamma {\sum }_{j=1}^N{W}_{ij}{Z}_{it}+{\nu }_i+{\phi }_t+{\epsilon }_{it} \end{gathered}$$

In Equations (8) and (9), ${RCS}_{it}$ represents a policy dummy variable of the RCS. The regression results presented in Table 5 demonstrate that the coefficients for both $FTZ$ and $W\ast FTZ$ maintain statistical significance, which confirms that after excluding the influence of the RCS, the implementation of the FTZ strategy still leads to a reduction in water pollution levels.

5.3.4. Change Methods

To ensure the reliability of the research findings, this paper relaxes the assumption conditions and employs the spatial autoregressive staggered difference-in-differences model (SAR-STA-DID) and the spatial error staggered difference-in-differences model (SEM-STA-DID) to further verify the spatial spillover effect of FTZs. As presented in Table 6, the regression results demonstrate consistently significant negative coefficients for the FTZ policy effect, spatial autoregressive term, and spatial error term, thereby confirming the robustness of our primary conclusions.

5.3.5. Replace Spatial Matrix

To overcome the errors in setting the weight matrix, this paper adopts two approaches to modify and replace the spatial matrix. Firstly, the inverse geographical distance matrix is standardized. Secondly, the economic distance is taken into account in terms of spatiality, and an economic–geographical composite spatial matrix is constructed. The calculation formula of the composite space matrix is as follows:

(10)$W_{ij}^{\prime }=\left\{\begin{array}{c}{\displaystyle \frac{1}{{|{pgdp}_i-{pgdp}_j|\ast d}_{ij}}},i\ne j\\ 0,i=j\end{array}\right.$

In Equation (10), $pgdp$ is the average per capita GDP of a city in the sample year; the meanings of other symbols are the same as those in Equation (3). The regression results in Table 7 show that the changes in the spatial matrix do not affect the direct and spatial inhibitory effects of FTZs on water pollution, passing the robustness test.

5.4. Analysis of Influence Mechanism

The purpose of this section is to delve deep into the internal mechanisms through which FTZs impact regional water pollution. Based on the theoretical analysis in Section 3, the FTZ strategy can reduce the level of water pollution via three channels: promoting industrial structure upgrading, accelerating technological innovation, and strengthening the environmental governance efforts. In the following, on the basis of the theoretical framework, a more in-depth empirical analysis of the relevant impact mechanisms will be carried out to uncover the profound causal relationship between the FTZ strategy and water pollution, thus providing a solid foundation for the formulation of targeted policies.

Table 8 presents the test results of the impact mechanism based on Equations (6) and (7). In Column (1), the coefficient of FTZ is significantly positive, indicating that the FTZ strategy can promote industrial structure upgrading. As shown in Column (2), after incorporating the mechanism variable of industrial structure upgrading into the benchmark model, the impacts of FTZs and industrial structure upgrading on the water pollution level are both significantly negative, suggesting that FTZs can improve the water pollution situation by promoting industrial structure upgrading, thus verifying Hypothesis 2. As elaborated in Section 3, FTZs can attract internationally renowned green energy and intelligent manufacturing enterprises to settle in, creating a “crowding out effect” on traditional water-intensive and high-pollution industries, thereby eliminating backward and highly polluting production methods. The dependence of overall economic activity on water resources decreases with the development of a high-end industrial structure, leading to a reduction in sewage discharge. Taking the Shanghai FTZ and Hebei FTZ as examples, the Shanghai FTZ has attracted green manufacturing enterprises such as the Tesla Gigafactory and Semiconductor Manufacturing International Corporation (SMIC), creating a “crowding-out effect” on traditional high-water-consuming industries, such as steel and chemical industries, and promoting the transformation of industries towards low pollution. The current water reuse rate is over 90%. However, the Hebei FTZ relies on ports to develop equipment manufacturing, resulting in the proportion of traditional heavy industry still being close to 30%. The improvement of water pollution is relatively limited and shows a marginally decreasing trend.

The coefficient of FTZ in Column (3) is significantly positive, verifying the realistic path of institutional easing promoting innovation emergence. From Column (4), after incorporating the mechanism variable of technological innovation into the benchmark model, the impacts of FTZs and technological innovation on the water pollution level are both significantly negative, indicating that FTZs can improve the water pollution situation by accelerating technological innovation, thus verifying Hypothesis 2. As elaborated in Section 3, the establishment of FTZs has significantly stimulated local innovation vitality. By accelerating the marketization process, it encourages enterprises to increase R&D investment and actively engage in innovation activities under more intense market competition and a more complete innovation incentive mechanism, especially in the fields of water treatment technology, clean production technology, and water-saving technology. There is a significant difference in the technological innovation paths between the Shenzhen FTZ and Xi’an FTZs. The Shenzhen FTZ relies on the innovative resources of the Greater Bay Area to accelerate research and development and achievement transformation in areas such as water treatment technology and clean production. Within its jurisdiction, Maozhou River has achieved Class V surface water standards by introducing membrane treatment technology. The latter has a weaker ability to gather innovative elements, and the supporting role of technology research in pollution control is relatively insufficient. The improvement effect of regional water quality is not as good as the former, and the compliance rate of industrial wastewater in the Wei River Basin within the region is only 90%.

In Column (5), the coefficient of FTZ is significantly positive, indicating that the FTZ strategy can strengthen the environmental governance efforts of governments. As shown in Column (6), after incorporating the mechanism variable of environmental governance efforts into the benchmark model, the impacts of FTZs and environmental governance efforts on the water pollution level are both significantly negative. This shows that FTZs can improve the water pollution situation by strengthening environmental governance efforts, thus verifying Hypothesis 2. As a testing ground for institutional innovation, FTZs have established an efficient emission reduction system through the expansion of the environmental governance investment scale and structural optimization, providing a demonstrative solution for the regional pollution prevention and control campaign. The comparison of governance models between the Hainan FTZ and Zhengzhou FTZs confirms the crucial role of government governance intensity in pollution control. The Hainan FTZ has innovatively established a “one vote veto” system for environmental protection. In 2023, the investment in environmental governance reached CNY 4.5 billion, and professional environmental protection enterprises were introduced through the Public–Private Partnership (PPP) model to achieve full collection and treatment of sewage, with a surface water quality rate of 96.9%. However, the Zhengzhou FTZ still focuses on logistics infrastructure, with insufficient investment in environmental governance. In 2023, ammonia nitrogen levels in the rivers within the zone still exceeded the standard.

6. Discussion

6.1. Interpretation of Findings

Firstly, this study finds that the construction of FTZs has a significant governance effect on water pollution. Theoretically, FTZs reduce the generation of water pollution from the source through the deep integration of green development concepts, institutional innovations such as emission rights trading, and strong support from green finance. Empirical results show that after the establishment of FTZs, the discharge of untreated sewage in local cities has significantly decreased, verifying the direct driving role of FTZs in water pollution control. Different from the current research on air pollution governance in FTZs [45], this study further expands the research field of FTZs’ environmental effects and confirms that FTZs also have positive effects on water pollution control. This finding provides more comprehensive empirical support for the green development positioning of FTZs, indicating that FTZs are not only highlands of economic openness but also innovative demonstration zones for environmental governance.

This study confirms three key mediating pathways through which FTZs affect water pollution. This conclusion is consistent with the existing literature that suggests that optimizing industrial structure, technological innovation, and government environmental governance can improve environmental quality [49,56,58].

One of the most innovative findings of this study is that free trade zones have a significant positive spatial spillover effect on water pollution control in surrounding areas, and the cumulative impact of this spillover effect exceeds the local direct impact. This finding is consistent with the theoretical expectation of spatial spillover effects in location-oriented policies [11] and also provides empirical evidence for regional collaborative governance of water pollution. This research achievement breaks through the limitation of previous studies only focusing on local effects and provides a new perspective for understanding the collaborative mechanism of regional environmental governance.

6.2. Policy Implications

6.2.1. Strengthen the Concept of Green Development and Institutional Innovation in FTZs

In the construction process of FTZs, the concept of green development should be further deepened and fully integrated into various aspects such as industrial planning, project approval, and enterprise operations. Continuously improve the trading market mechanisms for environmental rights such as emission permits and water rights. Through reasonable market pricing and trading rules, encourage enterprises to actively participate in water pollution control. At the same time, strengthen the construction of the regional environmental assessment system to ensure the scientific and fair nature of the assessment process.

6.2.2. Promote Industrial Structure Upgrading, Technological Innovation, and Environmental Governance in FTZs

Firstly, actively guide the industrial structure of FTZs to transform towards high-end and green development. Formulate targeted industrial support policies to encourage the clustered development of high-tech industries and environmental protection industries. Secondly, strengthen the construction of technological innovation platforms to promote cooperation among enterprises, universities, and research institutions. Jointly carry out research on key technologies for water pollution control. Thirdly, the government should further increase funding in the field of environmental governance and establish a dedicated environmental governance fund pool.

6.2.3. Construct a Regional Collaborative Governance Network of Water Pollution

Strengthen information sharing, technical exchange, and policy coordination between FTZs and surrounding areas by establishing a cross-regional cooperation mechanism for water pollution control. Build a regional water pollution control information platform to release real-time water quality monitoring data, pollution control project progress, and environmental policy updates from various regions, promoting communication and collaboration between regions. Develop regional unified environmental standards and regulatory policies to ensure that enterprises in the region follow the same standards in water pollution control, thereby forming a joint force for regional water pollution control.

6.3. Limitations and Prospects

Firstly, although using the discharge volume of untreated sewage in cities as an indicator has certain advantages, it may not fully reflect the complex situation of water pollution. Insufficient consideration of factors such as the types and concentrations of pollutants in wastewater leads to incomplete assessment of the degree of water pollution. In the future, more diverse indicators for measuring water pollution can be introduced. Besides the amount of sewage discharged, it is advisable to consider increasing the analysis of key pollutant indicators such as chemical oxygen demand, ammonia nitrogen, and heavy metal content in sewage. Combining satellite remote sensing monitoring data and online water quality monitoring data, a more comprehensive and accurate water pollution assessment system can be constructed.

Secondly, while the current study has validated the overall spatial spillover effect of FTZs on water pollution control, it has not conducted a detailed analysis of the specific mechanisms driving these spatial spillovers, nor has it quantitatively examined the spillover intensity and action pathways of each mechanism. Future research could further subdivide the four transmission mechanisms of “industrial structure—technological innovation—environmental regulation—factor flow” and construct a multi-dimensional collaborative model to quantify the spillover effect intensity and interactive effects of each mechanism across different spatial dimensions, providing precise theoretical support for regional collaborative pollution control.

7. Conclusions

This paper systematically reveals the impact of FTZs on water pollution through detailed theoretical mechanism deduction and multidimensional econometric testing. The specific conclusions are as follows:

Firstly, FTZs can improve the local water pollution situation. The positive and significant coefficient of the dependent variable in the STA-DID validates the corresponding hypothesis, and the results remain robust after a series of tests, including placebo tests and PSM-DID. Specifically, the establishment of FTZs reduces the discharge of untreated sewage in the local area by approximately 9.16 million tons per year.

Secondly, FTZs can generate positive spatial spillover effects, promoting water quality improvement in neighboring cities. The positive and significant coefficient of the indirect effect in the SDM-STA-DID method confirms the corresponding hypothesis, and the cumulative spillover effect exceeds the direct effect. Specifically, considering spatial effects, the establishment of FTZs reduces the discharge of untreated sewage in cities by approximately 21.23 million tons per year.

Thirdly, promoting industrial structure upgrading, accelerating technological in-novation, and strengthening government environmental governance serve as specific impact pathways and play a crucial mediating role in the influence of FTZs on water pollution.

Author Contributions

Conceptualization, methodology, data curation, supervision, validation, visualization, software, formal analysis, writing—original draft, and writing—review and editing, X.G. and J.S. (co-first authors); conceptualization, writing—review and editing, and supervision, X.Z.; supervision and visualization, G.D. and Y.L.; validation, J.Z. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Data will be made available on request.

Conflicts of Interest

The authors declare no conflicts of interest. Supporting entities had no role in the design of the paper, in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

Footnotes

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.

Figures and Tables

Figure 1 Influence framework (constructed based on the previous analysis).

Figure 2 Parallel trend test.

Figure 3 Placebo test results.

Table 1

Descriptive statistics.

Variable Type	Variables	Obs.	Mean	Std. Dev	Min	Max
Dependent variable	Water	5586	2135.143	5010.378	0.090	1.09×105
Independent variable	FTZ	5586	0.194	0.396	0.000	1.000
Control variable	Rate	5586	0.507	0.177	0.058	1.000
	Pop	5586	0.083	0.207	0.001	3.326
	Pgdp	5586	41,412.560	34,508.512	99.000	4.68 × 105
	Scale	5586	2.76 × 107	4.06 × 107	31,432	3.78 × 108
	Pipe	5586	1225.951	2205.890	5.000	43,249.360

Table 2

Results of benchmark regression.

	(1)	(2)	(3)	(4)
	STA-DID	STA-DID	SDM-STA-DID	SDM-STA-DID
FTZ	−888.1 ***	−916.6 ***	−878.3 ***	−894.7 ***
	(213.7)	(209.0)	(208.8)	(203.8)
W ∗ FTZ			−4211.9 **	−4955.8 ***
			(1798.7)	(1765.4)
City FixedTime FixedControl variables	YesYesNo	YesYesYes	YesYesNo	YesYesYes
N	5586	5586	5586	5586
R2	0.512	0.549	0.637	0.704

Note: significance levels: ** p < 0.05, and *** p < 0.01. Standard errors in parentheses. The following table also follows this instruction.

Table 3

Results of direct effect, indirect effect, and total effect.

	(1)	(2)
	SDM-STA-DID	SDM-STA-DID
Direct	−874.8 ***	−890.3 ***
	(209.4)	(204.7)
Indirect	−1074.0 **	−1232.9 ***
	(432.9)	(356.6)
Total	−1948.8 ***	−2123.1 ***
	(431.4)	(395.7)
City Fixed	Yes	Yes
Time Fixed	Yes	Yes
Control variables	No	Yes
N	5586	5586
R2	0.637	0.704

Note: significance levels: ** p < 0.05, and *** p < 0.01. Standard errors in parentheses.

Table 4

Test results of PSM-DID.

	(1)	(2)	(3)	(4)
	NN Matching	NN Matching	Caliper Matching	Caliper Matching
FTZ	−1002.6 ***	−981.3 ***	−998.1 ***	−977.0 ***
	(260.4)	(260.6)	(260.8)	(260.9)
City FixedTime FixedControl variables	YesYesNo	YesYesYes	YesYesNo	YesYesYes
N	4210	4210	4208	4208
R2	0.511	0.551	0.511	0.549

Note: significance levels: *** p < 0.01. Standard errors in parentheses.

Table 5

Test results of eliminating external interference.

	(1)	(2)	(3)	(4)
	STA-DID	STA-DID	SDM-STA-DID	SDM-STA-DID
FTZ	−888.1 ***	−868.7 ***	−878.3 ***	−848.4 ***
	(213.7)	(209.6)	(208.8)	(204.5)
W ∗ FTZ			−4211.9 **	−5186.7 ***
			(1798.7)	(1771.7)
City Fixed	Yes	Yes	Yes	Yes
Time Fixed	Yes	Yes	Yes	Yes
Control variables	No	Yes	No	Yes
N	5586	5586	5586	5586
R2	0.514	0.554	0.638	0.745

Note: significance levels: ** p < 0.05, and *** p < 0.01. Standard errors in parentheses.

Table 6

Test results of changing methods.

	(1)	(2)	(3)	(4)
	SAR-STA-DID	SAR-STA-DID	SEM-STA-DID	SEM-STA-DID
FTZ	−914.4 ***	−960.2 ***	−917.0 ***	−952.5 ***
	(210.0)	(204.5)	(208.3)	(203.5)
rho	−0.580 *	−1.126 ***
	(0.329)	(0.338)
lambda			−1.090 ***	−1.428 ***
			(0.363)	(0.379)
City Fixed	Yes	Yes	Yes	Yes
Time Fixed	Yes	Yes	Yes	Yes
Control variables	No	Yes	No	Yes
N	5586	5586	5586	5586
R2	0.621	0.685	0.633	0.692

Note: significance levels: * p < 0.10, and *** p < 0.01. Standard errors in parentheses.

Table 7

Test results of new spatial matrix.

	(1)	(2)	(3)	(4)
	Normalized Matrix	Normalized Matrix	Composite Matrix	Composite Matrix
FTZ	−924.8 ***	−954.3 ***	−1024.1 ***	−1049.3 ***
	(207.8)	(203.1)	(220.9)	(218.0)
W ∗ FTZ	−9059.9 ***	−8998.9 ***	−232.4 ***	−257.6 ***
	(2937.2)	(2871.2)	(49.92)	(49.31)
City Fixed	Yes	Yes	Yes	Yes
Time Fixed	Yes	Yes	Yes	Yes
Control variables	No	Yes	No	Yes
N	5586	5586	5586	5586
R2	0.646	0.708	0.625	0.705

Note: significance levels: *** p < 0.01. Standard errors in parentheses.

Table 8

Regression results of influence mechanism.

	(1)	(2)	(3)	(4)	(5)	(6)
	Industry	Water	Innovation	Water	Governance	Water
FTZ	0.441 ***	−1695.6 ***	82.91 ***	−258.1 **	263.8 ***	−243.3 **
	(0.0109)	(209.2)	(24.81)	(107.0)	(29.68)	(108.1)
Industry		−1166.4 ***
		(245.1)
Innovation				−0.303 ***
				(0.0686)
Governance						−0.151 ***
						(0.0575)
City Fixed	Yes	Yes	Yes	Yes	Yes	Yes
Time Fixed	Yes	Yes	Yes	Yes	Yes	Yes
Control variables	Yes	Yes	Yes	Yes	Yes	Yes
N	4224	4224	4224	4224	4224	4224
R2	0.292	0.551	0.477	0.562	0.698	0.577

Note: Due to the serious absence of data on mediator variables, the data from 2003 to 2007 are deleted. Significance levels: ** p < 0.05, and *** p < 0.01.

Appendix A

Table A1

List of 266 cities.

The Abbreviation of the City Name
sjz	as	hh	mas	fz	ny	sg	hc	xa
ts	fs	sh	hb	sr	sq	sz	hk	tc
qhd	bx	nj	tl	jn	xy	zh	sy	bj
hd	dd	wx	aq	qd	zk	st	cd	xy
xt	jz	xz	hs	zb	zmd	fs	zg	wn
bd	yk	cz	cz	zz	wh	jm	pzh	ya
zjk	fx	sz	fy	dy	hs	zj	zz	hz
cd	ly	nt	sz	yt	sy	mm	dy	yl
cz	pj	lyg	la	wf	yc	zq	my	ak
lf	tl	ha	zz	jn	ez	hz	gy	sl
hs	cy	yc	cz	ta	jm	mz	sn	lz
ty	hld	yz	xc	wh	xg	sw	nj	jyg
dt	cc	zj	fz	rz	jz	hy	ls	jc
yq	jl	tz	xm	ly	hg	yj	nc	by
cz	sp	sq	pt	dz	xn	qy	ms	ts
jc	ly	hz	sm	lc	sz	dz	yb	ww
sz	th	nb	qz	bz	cs	zs	ga	zy
jz	bs	wz	zz	hz	zz	cz	dz	pl
yc	sy	jx	np	zz	xt	yf	ya	jq
xz	bc	hz	ly	kf	hy	nn	bz	xn
lf	heb	sx	nd	ly	sy	lz	zy	yc
hhht	qqhe	jh	nc	pds	yy	gl	gy	szs
bt	jx	zz	jdz	ay	cd	wz	lps	wz
wh	hg	zs	px	hb	zjj	bh	zy	gy
cf	sys	tz	jj	xx	yy	fcg	as	wlmq
tl	dq	ls	xy	jz	cz	qz	km	klmy
eeds	yc	hf	yt	zy	yz	gg	qj
hlbe	jms	wh	gz	xc	hh	yl	yx
sy	qth	bb	ja	zh	ld	bs	bs
dl	mdj	hn	yc	smx	gz	hz	zt

References

1. Zhou, X.F.; Cao, G.Z.; Yu, F.; Liu, Q.; Ma, G.; Yang, W. Risk zoning of acute water pollution in the Yangtze River Economic Belt. Acta Sci. Circumst.; 2020; 1, pp. 334-342.

2. Babuji, P.; Thirumalaisamy, S.; Duraisamy, K.; Periyasamy, G. Human Health Risks due to Exposure to Water Pollution: A Review. Water; 2023; 15, 2532. [DOI: https://dx.doi.org/10.3390/w15142532]

3. Diao, X.S.; Agapi, S. A dynamic evaluation of the effects of a free trade area of the Americas—An intertemporal, global general equilibrium model. J. Econ. Integr.; 2001; 1, pp. 21-47. [DOI: https://dx.doi.org/10.11130/jei.2001.16.1.21]

4. Jiang, R.C.; Yu, D.F.; Zhai, Q.; Wang, D.; Niu, Z.Y. On the effect of China (Shanghai) pilot free trade zone on headquarters economic development in Shanghai. Foreign Econ. Manag.; 2014; 4, 65–71+80

5. Li, Z.L.; Liu, D. Research on the “quality effect” of the construction of China’s pilot free trade zone. Economist; 2021; 9, pp. 58-68.

6. Hu, Y.N.; Cheng, H.F. Water pollution during China’s industrial transition. Environ. Dev.; 2013; 1, pp. 57-73. [DOI: https://dx.doi.org/10.1016/j.envdev.2013.06.001]

7. Wu, J.; Wang, Q. Clean production and wastewater treatment technologies to reduce water pollution in the manufacturing sector. J. Clean. Prod.; 2019; 227, pp. 1015-1023.

8. Wang, M.; He, Y.; Zhou, J.; Ren, K. Evaluating the effect of Chinese environmental regulation on corporate sustainability performance: The mediating role of green technology innovation. Int. J. Environ. Res. Public Health; 2022; 11, 6882. [DOI: https://dx.doi.org/10.3390/ijerph19116882]

9. Li, S.J.; Zhao, T.R. Has the pilot free trade zone promoted the upgrading of industrial structure: An empirical analysis based on China (Shanghai) pilot free trade zone. J. Cent. Univ. Financ. Econ.; 2019; 8, pp. 118-128.

10. Jiao, Y.X.; Yang, H.E.; Wang, Y.; Yu, H.L. Research on the impact of construction of pilot free trade zones on corporate R&D investment: An analysis based on the moderating effect of marketability process. Sci. Res. Manag.; 2024; 4, pp. 111-120.

11. Kline, P.; Moretti, E. People, Places, and Public Policy: Some Simple Welfare Economics of Local Economic Development Programs. Annu. Rev. Econ.; 2014; 1, pp. 629-662. [DOI: https://dx.doi.org/10.1146/annurev-economics-080213-041024]

12. LeSage, J.P.; Dominguez, M. The importance of modeling spatial spillovers in public choice analysis. Public Choice; 2012; 1, pp. 525-545. [DOI: https://dx.doi.org/10.1007/s11127-010-9714-6]

13. She, Y.; Liu, Y.B. The effectiveness of “bottom-up” environmental policies: A comparative analysis of heterogeneity of the River Chief System Policy in the Yangtze River Economic Belt. Resour. Sci.; 2023; 7, pp. 1139-1152. [DOI: https://dx.doi.org/10.18402/resci.2023.06.04]

14. Du, J.; Li, B. Assessing the impact of China’s river chief system on enterprise pollution discharge. Front. Public Health; 2023; 11, 1268473. [DOI: https://dx.doi.org/10.3389/fpubh.2023.1268473]

15. Du, H.J.; Deng, Q.Z. River chief governance system: Policy instruments, hydro-engineering and system governance effectiveness. China Popul. Resour. Environ.; 2024; 2, pp. 201-212.

16. Ma, J.Q.; Le, Z.; Xia, Y. Research on the influence of lake chief system on water environment treatment effect: Taking Wuhan city as an example. Resour. Environ. Yangtze Basin; 2022; 5, pp. 1125-1136.

17. Jing, S.W.; Zhang, J. Can Xin’anjiang river basin horizontal ecological compensation reduce the intensity of water pollution?. China Popul. Resour. Environ.; 2018; 10, pp. 152-159.

18. Davis, J.C.; Paudel, K.P.; Rupasingha, A. Role of income and policy in reducing water pollution: Evidence from the Mississippi river basin healthy watersheds initiative. Appl. Econ. Perspect. Policy; 2024; 2, 1. [DOI: https://dx.doi.org/10.1002/aepp.13417]

19. Shao, S.; Liu, L.W. Evaluating the effect of water pollution governance policy in China: Evidence from the water ecological civilization city construction pilot. Reform; 2023; 2, pp. 75-92.

20. Lyu, L.J.; Kan, D.X.; Yao, W.Q.; Huang, W.C. Has China’s pilot policy of water ecological civilization city construction reduced water pollution intensity?. Land; 2022; 11, 1999. [DOI: https://dx.doi.org/10.3390/land11111999]

21. Yang, G.S.; Hu, M.L.; Yang, F.L. Efficiency and equity balanced water pollutant emission aggregate allocation for Jiangsu province. Resour. Environ. Yangtze Basin; 2024; 4, pp. 833-843.

22. Li, W.B.; Yang, Y.W. Flattening reform of environmental monitoring and enforcement agencies and transboundary water pollution. China Popul. Resour. Environ.; 2023; 4, pp. 137-146.

23. Li, Y.R.; Li, L.; Zhai, H.Y. Outgoing audit of natural resource assets and water pollution control: Based on empirical data from water quality monitoring in major river basins in China. Jiangxi Soc. Sci.; 2022; 5, pp. 90-100.

24. Chen, J.; Tian, G.L.; Li, J.W.; Xu, H.J. Research on the outgoing audit and evaluation of water resource assets of leadership cadres in city y. Sustainability; 2023; 16, 12535. [DOI: https://dx.doi.org/10.3390/su151612535]

25. Zhou, K.; Yin, Y.; Chen, Y.F. Driving factors and scale effects of water pollutant discharge in the urban agglomeration. Acta Geogr. Sin.; 2022; 9, pp. 2219-2235.

26. Li, J.Q.; Shen, Z.Y.; Liu, G.W.C.; Jin, Z.T.; Liu, R.Z. The effect of social economy-water resources-water environment coupling system on water consumption and pollution emission based on input-output analysis in Changchun city, China. J. Clean. Prod.; 2023; 423, 138719. [DOI: https://dx.doi.org/10.1016/j.jclepro.2023.138719]

27. Luo, Z.; Qi, B.C. The effects of environmental regulation on industrial transfer and upgrading and banking synergetic development: Evidence from water pollution control in the Yangtze River basin. Econ. Res. J.; 2021; 2, pp. 174-189.

28. Zhang, M.Y.; Li, F. Can the establishment of free trade zone promote the high-quality development of regional economy: Based on the analysis of free trade zones in central and western China. West Forum Econ. Manag.; 2022; 06, pp. 57-68.

29. Zhang, T.T.; Wang, W.X. The spatial spillover effects of establishing free trade pilot zones on FDI: Empirical evidence from China. Heliyon; 2023; 10, e20587. [DOI: https://dx.doi.org/10.1016/j.heliyon.2023.e20587]

30. Ren, Z.P.; Tian, S.T.; Shi, N. The effects of pilot free trade zones on the regional advantages of and complementarity between four regions in China: Evidence from the Dendrinos-Sonis model. China Soft Sci.; 2016; 11, pp. 175-183.

31. Peng, Y.; Yang, Z.Y. Does the construction of the pilot free trade zones bring regional radiation effects empirical research based on the Yangtze River delta pearl river delta, and Beijing-Tianjin-Hebei regions. J. Int. Trade; 2020; 9, pp. 65-80.

32. Wang, L.H.; Liu, Z.H.; Shi, H.L. The impact of the pilot free trade zone on regional financial development. China World Econ.; 2022; 5, pp. 154-184. [DOI: https://dx.doi.org/10.1111/cwe.12441]

33. Liu, B.L.; Wang, Y. Innovative performance promotion effect of free trade zone: Evidence from the quasi-experiment of the Shanghai free-trade zone. Res. Econ. Manag.; 2018; 09, pp. 65-74.

34. Xu, S.T.; Shen, R.H.; Zhang, Y.H.; Cai, Y.F. Fostering regional innovation efficiency through pilot free trade zones: Evidence from China. Econ. Anal. Policy; 2024; 81, pp. 356-367. [DOI: https://dx.doi.org/10.1016/j.eap.2023.12.004]

35. Dai, X.; Xu, H.F.; Zeng, L.H. Does innovation in pilot free trade zone system improve export resilience?. Reform Econ. Syst.; 2023; 4, pp. 23-32.

36. Xiang, H.J.; He, K.; Yu, Y. Establishment of FTZ, trade development and capital flows based on the perspective of Shanghai free trade zone. J. Financ. Res.; 2016; 10, pp. 48-63.

37. Yang, S.P.; Inmaculada, M. A panel data analysis of trade creation and trade diversion effects: The case of ASEAN–China free trade area. China Econ. Rev.; 2014; 29, pp. 138-151. [DOI: https://dx.doi.org/10.1016/j.chieco.2014.04.002]

38. Zhang, Y.W. Research on the impact of the establishment of free trade zones on the total factor productivity of enterprises: Based on the mediating role of financing constraints. Mod. Bus.; 2021; 23, pp. 61-68.

39. Jiang, Y.F.; Wang, H.Y.; Liu, Z.K. The impact of the free trade zone on green total factor productivity: Evidence from the shanghai pilot free trade zone. Energy Policy; 2021; 148, 112000. [DOI: https://dx.doi.org/10.1016/j.enpol.2020.112000]

40. Yu, Z.X.; Zhang, J.; Dong, J.M.; Yang, Y. Does comprehensive bonded zone policy improve the transformation and upgrading of exporters?. Stat. Res.; 2022; 8, pp. 53-71.

41. Fang, Y.L.; Liu, J.X. Can the establishment of pilot free trade zones promote enterprise innovation? Evidence from GEM listed companies. Stud. Int. Financ.; 2021; 9, pp. 25-33.

42. Jiang, Y.; Sui, G.J.; Yang, Y.C. Mechanisms and paths of the construction of the pilot free trade zone to promote the resilience of industrial chain and supply chain: Taking Guangdong-Hong Kong-Macao Greater Bay Area as an example. Intertrade; 2023; 6, pp. 55-63.

43. Ba, S.S.; Chai, H.R.; Fang, Y.L.; Wang, B. Does the establishment of the pilot free trade zone improve the efficiency of financial services for the real economy?. World Econ. Stud.; 2021; 12, 3–21+132

44. Chauffour, J.P.; Maur, J.C. Preferential trade agreement policies for development: A handbook. World Bank Publ.; 2011; 5, pp. 662-687.

45. Hu, Z.Y.; Zhou, J.K.; Li, Y. Has the establishment of free trade zones improved the atmospheric conditions?. China Popul. Resour. Environ.; 2022; 2, pp. 37-50.

46. Zhi, P.Y.; Huang, L.Q.; Chen, Q. Construction of pilot free trade zone and regional industrial structure transformation and upgrading empirical analysis based on panel data of 286 cities in China. South China J. Econ.; 2021; 4, pp. 37-54.

47. Yu, Y.Z.; Sun, P.B.; Xuan, Y. Do constraints on local governments’ environmental targets affect industrial transformation and upgrading?. Econ. Res. J.; 2020; 8, pp. 57-72.

48. Xiang, C.X.; Chen, Z.M.; Zheng, X.Y. Influence of environmental externalities on export structure and trade policies. China Popul. Resour. Environ.; 2021; 6, pp. 45-56.

49. Zhu, B.S.; Guo, R.; Tang, Y.E.; Peng, K.M.; Huang, X.F. Upgrading the industrial structure for optimizing water-energy-carbon nexus in regional trade network. J. Clean. Prod.; 2024; 469, 143231. [DOI: https://dx.doi.org/10.1016/j.jclepro.2024.143231]

50. Lata, S.; Siddharth,. Sustainable and eco-friendly approach for controlling industrial wastewater quality imparting succour in water-energy nexus system. Energy Nexus; 2021; 3, 100020. [DOI: https://dx.doi.org/10.1016/j.nexus.2021.100020]

51. Xue, J.; Weng, Z.X.; Liu, T.T.; Li, H.S. Research on fexible science and technology governance mechanisms for ecological protection and high-quality development in the Yellow River Basin. Environ. Prot.; 2024; 7, pp. 38-41.

52. Ke, Y.M.; Liu, W.; Wang, J.X.; Liu, F.T. How to promote the sustainable development of strategic emerging industries: Evidence from Guangzhou. J. Clean. Prod.; 2023; 428, 139522. [DOI: https://dx.doi.org/10.1016/j.jclepro.2023.139522]

53. Wang, B.B.; Qi, S.Z. The effect of market-oriented and command-and-control policy tools on emissions reduction innovation: An empirical analysis based on China’s industrial patents data. China Ind. Econ.; 2016; 6, pp. 91-108.

54. Qi, S.Z.; Lin, S.; Cui, J.B. Do environmental rights trading schemes induce green innovation? Evidence from listed firms in China. Econ. Res. J.; 2018; 12, pp. 129-143.

55. Mohsen, A.; Mostafa, E.A.; Milad, H.C. Performance of the firms in a free-trade zone: The role of institutional factors and resources. Eur. Manag. Rev.; 2019; 2, pp. 363-378.

56. Siddique, I. Sustainable water management in urban areas: Integrating innovative technologies and practices to address water scarcity and pollution. Pharm. Chem. J.; 2021; 1, pp. 172-178. [DOI: https://dx.doi.org/10.2139/ssrn.4883898]

57. Liu, J.; Chen, L. Innovative technologies for water pollution control and treatment in urban areas. Water Res.; 2017; 120, pp. 228-243.

58. She, Y.; Liu, Y.; Jiang, L.; Yuan, H. Is China’s river chief policy effective? Evidence from a quasi-natural experiment in the Yangtze River Economic Belt, China. J. Clean. Prod.; 2019; 220, pp. 919-930. [DOI: https://dx.doi.org/10.1016/j.jclepro.2019.02.031]

59. Shen, K.R.; Jin, G.; Fang, X. Does environmental regulation cause pollution to transfer nearby. Econ. Res. J.; 2018; 52, pp. 44-59.

60. Li, J.; Shi, X.; Wu, H.; Liu, L. Trade-off between economic development and environmental governance in China: An analysis based on the effect of river chief system. China Econ. Rev.; 2019; 60, 101403. [DOI: https://dx.doi.org/10.1016/j.chieco.2019.101403]

61. Yang, Y.C.; Niu, X.T. Impact of social trust on the effectiveness of environmental regulations on carbon emission reduction: Based on panel data from 281 prefecture-level cities in China. China Popul. Resour. Environ.; 2023; 4, pp. 82-92.

62. Zhang, G.; Zhang, P.; Zhang, Z.G.; Li, J. Impact of environmental regulations on industrial structure upgrading: An empirical study on Beijing-Tianjin-Hebei region in China. J. Clean. Prod.; 2019; 238, 117848. [DOI: https://dx.doi.org/10.1016/j.jclepro.2019.117848]

63. Bronzini, R.; Piselli, P. The impact of R&D subsidies on firm innovation. Res. Policy; 2016; 2, pp. 442-457.

64. Wang, Y.; Feng, J.H. Research on green bond promoting green innovation of enterprises. J. Financ. Res.; 2022; 6, pp. 171-188.

65. Ge, T.; Hao, X.; Li, J. Effects of public participation on environmental governance in China: A spatial Durbin econometric analysis. J. Clean. Prod.; 2021; 25, 321. [DOI: https://dx.doi.org/10.1016/j.jclepro.2021.129042]

66. Fan, W.; Yan, L.; Chen, B.; Ding, W.; Wang, P. Environmental governance effects of local environmental protection expenditure in China. Resour. Policy; 2022; 77, 102760. [DOI: https://dx.doi.org/10.1016/j.resourpol.2022.102760] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36569596]

67. Adina, D.; Ada-Iuliana, P.; Dumitru-Cristian, O. Government expenditure on environmental protection and environmental performance: Evidence from EU countries. Eur. Financ. Resil. Regul.; 2023; 2023, pp. 84-96.

68. Ding, R.Z.; Li, X.M. Spatial effect of provincial-level free trade zone pilot policy on urban economic growth. Econ. Geogr.; 2024; 2, pp. 21–30+42.

69. Ma, Y.Y.; Yao, W.Y.; Jiang, L.; Xue, Y.W. Impact of Chin’′s green finance reform and innovation pilot zone policies on urban pollution and carbon emission reduction and their mechanisms. China Popul. Resour. Environ.; 2024; 61, pp. 45-55.

70. Shao, S.; Ge, L.M.; Zhu, J.L. How to achieve the harmony between humanity and nature: Environmental regulation and environmental welfare performance from the perspective of geographical factors. J. Manag. World; 2024; 08, pp. 119-146.

71. Xu, S.; Liu, T.Z.; Liu, Y.H. Spillover effect of green finance on co, emissions from a spatial perspective: The Yangtze River economic belt as an example. Resour. Environ. Yangtze Basin; 2024; 6, pp. 1313-1324.

72. Miloš, K. Climate change in the EU: Analysis by clustering and regression. Serbian J. Manag.; 2023; 1, pp. 111-132. [DOI: https://dx.doi.org/10.5937/sjm18-43601]

73. Jiang, T. Mediating effects and moderating effects in causal inference. China Ind. Econ.; 2022; 5, pp. 100-120.

74. Bardaka, E.; Delgado, M.S.; Florax, R.J.G.M. A spatial multiple treatment multiple outcome difference-in-differences model with an application to urban rail infrastructure and gentrification. Transp. Res. Part A Policy Pract.; 2019; 121, pp. 325-345. [DOI: https://dx.doi.org/10.1016/j.tra.2019.01.028]

75. Li, G.Q.; He, R.W. PM2.5 Pollution and health expenditures: Time lag effects and spatial spillovers. J. Saf. Environ.; 2019; 1, pp. 326-336.

76. Alzahrani, F.; Collins, A.R.; Elham, E. Drinking water quality impacts on health care expenditures in the United States. Water Resour. Econ.; 2020; 32, 100162. [DOI: https://dx.doi.org/10.1016/j.wre.2020.100162]

77. Shen, K.R.; Jin, G. The policy effects of local governments’ environmental governance in China: A study based on the evolution of the “River-Director” system. Soc. Sci. China; 2018; 5, 92–115+206