Background

After the 2008 financial crisis, trade protectionism emerged in various countries, and regional bilateral and multilateral free trade agreements continued to emerge, such as the TPP, TTIP, and TISA. The signing and development of these regional trade agreements have had a significant impact on global trade rules and standards, directly or indirectly affecting China’s comparative advantage, market access, and international discourse power, making China face more severe challenges in integrating into globalization strategies. At the same time, China is also facing urgent practical problems such as the gradual decline of traditional factor dividends, difficulties in industrial structure adjustment and upgrading, and slow progress in economic system reform. This makes it difficult for the country’s administrative management system, economic freedom, and business environment to adapt to the current high-standard economic and trade rules in the world.

Given the aforementioned domestic and international background, the State Council issued the Overall Plan for China Pilot FTZ in September 2013, establishing the first domestic FTZ in Shanghai, actively connecting with new rules and standards for international trade and investment, and using this to force domestic market economy system reform. In order to promote pilot experience and amplify the institutional dividend effect of FTZ, the State Council has successively approved the establishment of FTZ in Guangdong, Tianjin, Fujian, Liaoning, Zhejiang, Henan, Hubei, Chongqing, Sichuan, Shaanxi, Hainan, Shandong, Jiangsu, Guangxi, Hebei, Yunnan, Heilongjiang, Beijing, Hunan, and Anhui. At this point, China has a total of 22 FTZ, covering two-thirds of the country’s provinces and cities, forming a new pattern of “1+3+7+1+6+3” reform and opening up.

As of July 2021, a total of 278 institutional innovation achievements at the national level and approximately 1400 replicable and promotable institutional innovation achievements at the local level have been summarized in 22 FTZ, involving government function transformation, investment and trade liberalization and facilitation, financial openness and innovation, the rule of law environment, and regulatory systems. Since the establishment of FTZ, China has achieved remarkable results in economic system reform and innovation and has made great achievements in optimizing the business environment. According to the Global Business Environment Report released by the World Bank over the years, China’s business environment index has jumped from 96th place in 2013 to 31st place in 2022.

Theoretical Analysis

The establishment of FTZ is an important policy for the country to adapt to various changes and contradictions, deepen reform, and expand its opening up in the new situation. With the support of the local government, many FTZs have innovated their institutional mechanisms from multiple dimensions in order to build them into internationally competitive parks. Firstly, the establishment of FTZ can fully leverage reform autonomy, carry out institutional innovation in areas such as trade, finance, and the legal system, attract high-quality enterprises through relevant policies, and promote the optimization and upgrading of urban industrial structures. Industrial structure upgrading is a key path to curbing environmental pollution. On the one hand, the establishment of FTZ generates a competitive effect through efficient resource allocation and increased entry barriers, promoting the survival of the fittest among enterprises, eliminating backward enterprises that cause serious environmental pollution, and suppressing the aggravation of environmental pollution. On the other hand, the establishment of FTZ attracts a large number of high-tech enterprises to settle in the park by absorbing foreign investment. At the same time, it guides the transformation of park enterprises from resource intensive to technology intensive, promotes the upgrading of enterprise production methods through innovative green technologies, and further reduces environmental pollution. In terms of the impact of industrial structure upgrading on environmental pollution, Cole et al. (2003) argue that capital intensive production is highly correlated with environmental pollution [11]. Lin (2017) found that the reduction of trade barriers significantly reduced the carbon dioxide emissions of enterprises, which was mainly due to the reduction of coal use intensity and cleaner production processes [12]. In addition, some scholars have found that the larger the proportion of the secondary industry, the higher the degree of environmental pollution, while the tertiary industry can alleviate the deepening of environmental pollution. The industrial structure is closely related to the degree of environmental pollution [13, 14]. In summary, it can be considered that upgrading the industrial structure is beneficial for improving the environmental pollution situation.

Secondly, in the process of constructing FTZ, the degree of openness to the outside world gradually increases, the degree of marketization also increases, and the allocation of resource elements is further improved, among which the allocation of environmental resource elements is also optimized. In the process of continuously optimizing resource allocation and guided by market mechanisms, enterprises change their production methods, promote technological innovation, and provide the impetus for the green development of FTZ and regional sustainable development. According to Porter’s hypothesis, incentivizing environmental regulatory policies will encourage enterprises to engage in technological innovation, which will increase their productivity, compensate for the cost losses caused by environmental governance, and enhance their competitiveness. The coordination role between government policies and environmental protection can be fully utilized. The government has formulated multiple preferential policies, and the institutional mechanism innovation of FTZ has attracted enterprises to increase their technological innovation efforts, driving local enterprises to improve emission reduction technologies and reduce environmental pollution [15, 16]. In summary, it can be considered that technological innovation can suppress urban environmental pollution. Based on this, this article proposes the following hypothesis:

Hypothesis 1: The establishment of FTZ reduces urban environmental pollution through two paths: upgrading industrial structures and technological innovation.

The establishment of FTZ has reduced trade barriers between regions, effectively promoted the free flow of factors between regions, and made trade activities in different regions no longer limited to their own regions. It has strengthened trade and economic connections between regions and, to some extent, changed the regional development pattern. At the same time, the development of FTZ not only promotes the cities they are located in, but also serves as a demonstration and driving force for other regions, thus their policy effects have a certain regional radiation power [17, 18]. Chen and Hall (2011) believe that the convenience of factor flow between regions is an important factor affecting industrial upgrading, thereby promoting technology spillovers [19]. Wang et al. (2019) measured the international competitiveness of China’s manufacturing industry and found that the expansion of the FTZ network can promote the growth of the country’s manufacturing output value, and the increase in foreign trade dependence can enhance the competitiveness of the country’s manufacturing industry [20]. However, based on different association rules, the establishment of FTZ will have different spatial effects on the manufacturing industries of different neighboring countries or regions. Jiang et al. (2021) studied the policy effects of FTZ establishment on air pollution and found that pilot cities can not only improve air pollution through industrial structure upgrading and technological innovation, but also generate spatial spillover effects to improve the atmospheric environment of neighboring cities [21]. According to existing research, the establishment of FTZ has spatial spillover effects on the economic development, industrial development, and other aspects of surrounding cities. Therefore, the inhibitory effect of FTZ establishment on environmental pollution should not be limited to this region, and the environmental pollution of surrounding cities may also be improved by the establishment of FTZ. Therefore, this article proposes the following hypothesis:

Hypothesis 2: There is a spatial spillover effect of the establishment of FTZ on urban environmental pollution.

Data Source

Our dataset comprises multiple sources, mainly including the urban environmental pollution index, the establishment of FTZ, and other data at the city level.

Urban environmental pollution index: To scientifically reflect the environmental pollution situation in various cities, we calculate the comprehensive environmental pollution index of each city from three dimensions: industrial wastewater, sulfur dioxide, and smoke dust emissions, using the entropy method, and correct outliers to more accurately reflect the regional environmental pollution situation. Considering the availability of data and to ensure the consistency of the data sample, this article has determined the data sample period to be from 2010 to 2021.

Establishment of FTZ: The independent variable is a dummy variable of whether a city has established an FTZ. If the city has established an FTZ, the value of did is assigned to 1, otherwise it is 0. Post is a policy time dummy variable. If the city established an FTZ in 2015, the value of Post is 1 in 2015 and beyond; otherwise, it is 0. Meanwhile, considering that the FTZ established from 2018 to 2019 has not yet produced significant economic and environmental effects during the sample period, the selected sample for the treatment group is 22 cities that established FTZ from 2013 to 2017, with other cities as the control group.

Mechanism variable: Industrial structure upgrading, using the ratio of the added value of the tertiary industry in each city to the sum of the added values of the primary and secondary industries to represent the overall upgrading status of the city’s industrial structure. The level of technological innovation is characterized by the number of patent applications in each city in the current year.

Other city-level data mainly include the economic development level, the proportion of the secondary industry to GDP, FDI, city size, and urbanization rate. These data mainly come from the website of the National Bureau of Statistics and the China City Statistical Yearbook.

Table 1 reports the summary statistics of the above dependent variable, independent variable, control variables, and mechanism analysis variables.

Empirical Strategy

The most commonly used method for evaluating policy effects is the DID method. Due to the differences in the timing of the establishment of each trade zone, in order to better examine the impact of the establishment of FTZ on the environment, this article chooses a staggered DID.

In Eq. (1), pollution_it is the outcome variable, which is represented by the urban environmental pollution index; did_it denotes the independent variable, which is a dummy variable used to describe FTZ. It will be coded 1 in the current and subsequent years if city i is piloted in year t and 0 otherwise. X_it denotes a series of control variables that may affect environmental pollution, including economic development level, the proportion of the secondary industry to GDP, FDI, city size, and urbanization rate. year_t and city_t are year fixed effects and city fixed effects, respectively, and ε_it is the error term. The coefficient β₁ on the post-visit dummy captures the treatment effect of the FTZ on urban environmental pollution.

Baseline Results

Starting with the staggered DID of the baseline, we estimate Eq. (1) using the urban environmental pollution index from 2010 to 2021 as the dependent variable. To ensure the accuracy of the empirical results, Table 2 presents our baseline results obtained by not adding control variables and adding control variables. Estimates are separately reported in Columns (1) and (2). The estimated coefficients of interest are highly significant. The results in Table 2 show that the establishment of FTZ has reduced urban environmental pollution as measured by the urban environmental pollution index. Specifically, in the results of Column (2), the estimated coefficient is -0.400, which means that after controlling the relevant control variables and fixed effects, the cities piloted by the FTZ have significantly lower pollution emissions than those cities not piloted.

Parallel Trend Test

To prove the reliability of our conclusion, we conducted parallel trend tests. The event study provides a reference for our analysis, which can determine the dynamic trends before and after the occurrence of policies [22]. The test equation is set up as follows:

In Eq. (2), did_ti0+k denotes the FTZ establishment. Specifically, t_i0 denotes the time spent by FTZ in different cities, and k denotes the years before and after the city implemented the FTZ. If t – t_i0≤–5, then did_ti0–5 = 1, otherwise, it is 0. If t – t_i0 = k (k = –5, –4, –3, –2, 0, 1, 2, 3), did_ti0+k = 1; otherwise, it is 0. If t – t_i0≥3, then did_ti0+3 = 1, otherwise, it is 0. This study sets k = –1 as the base period, and the coefficient β_k reflects the influence of FTZ on urban environmental pollution in the k-th year.

Fig. 1 displays the point estimate of Eq. (2) and the results at the 95% confidence interval. The figure lends strong support to the parallel-trend assumption, which implies that the processing and control groups meet the requirements for counterfactual estimation. Meanwhile, there is a gradual and significant reduction in urban environmental pollution after the FTZ establishment, which suggests that the FTZ has a long-term role in reducing urban environmental pollution.

Placebo Test

We also conducted a placebo test. Specifically, we counted the number of cities implementing FTZ from 2010 to 2021. Among them, there were 22 cities implemented. Therefore, in order to construct a counterfactual estimation, we randomly select years from the sample period and label them as t. Then, 22 cities were randomly selected in year t as the cities implementing FTZ. Thus, randomly assigned FTZ data is used for the placebo test. To maintain the robustness of the placebo results, we repeated them 500 times.

Fig. 2 shows the distribution of estimated values and baseline estimates for 500 runs. We found that our estimated results are outside the entire distribution, indicating that our results are not caused by unobservable factors.

Robustness Test

To eliminate some other concerns, we conducted some robustness tests.

First, we tested the robustness of the urban environmental pollution index. Considering local government officials may manipulate environmental pollution data to expect a higher chance of being promoted [23], we used PM 2.5 data for robustness testing. This data is sourced from the Atmospheric Composition Analysis Group at the University of Washington, St. Louis, and is a satellite remote sensing image formed by combining satellite images with ground monitoring station data with a resolution of 1 km. Currently, this data has been used in many haze studies [24]. According to Eq. (1), we obtain the result in Column (1) of Table 3.

Secondly, we examined the lagging effect of FTZ establishment. Due to a certain lag in the implementation of the policy, which may result in an underestimate of the baseline results, we have delayed the FTZ establishment by one period and re-estimated Eq. (1). We obtain the result in Column (2).

Third, we narrowed down the sample size. Specifically, we excluded large cities with higher administrative powers. Due to the fact that those cities have higher administrative levels, richer market resources, and more complete social systems, there may be differences in energy use and consumption. Therefore, we deleted the city data. We obtain the result in Column (3).

Fourth, in order to exclude the impact of other policies, we separately controlled the relevant policies for the period from 2010 to 2021. Specifically, we mainly controlled for green finance policy and innovative city policy. We obtain the result in Columns (4) and (5).

Finally, considering the possibility of heterogeneity treatment effects in grouping and time dimensions, there may be some bias in the estimation of staggered DID; therefore, we used the more robust estimators proposed by Sun and Abraham (2020) and Callaway and Sant Anna (2021) for estimation [25, 26]. Among them, Sun and Abraham’s (2020) estimated results are shown in Column (6), while Callaway and Sant Anna’s (2021) estimated results are shown in Column (7) [25, 26]. The regression results show that even after considering the effects of heterogeneity treatment, a more robust estimator is still significant.

Fig. 1.

Parallel trend test of FTZ on urban environmental pollution.

Fig. 2.

Counterfactual estimates of FTZ.

As expected, all results are negatively significant in Table 3. Therefore, we believe that our results are robust.

Analysis of Heterogeneity

The baseline regression is a study conducted on 285 prefecture level and above cities in China as a whole, and the estimated results indicate that the establishment of FTZ has a significant inhibitory effect on environmental pollution. However, the differences in economic development status, industrial development direction, and resource factors among different cities may lead to different effects of the establishment of FTZ on environmental pollution. Therefore, 285 cities were divided into two types of samples: resource-based cities and non resource-based cities, central cities and non central cities, to explore the regional heterogeneity of the impact of FTZ establishment on environmental pollution.

Columns (1) and (2) in Table 4 show the regression results after dividing the sample into resource-based cities and non resource-based cities. Observing the data from both columns, it can be seen that the regression coefficient of FTZ establishment on environmental pollution in non resource-based cities is -0.319, which is significant at the 1% level. This indicates that FTZ establishment has a significant improvement effect on the environmental pollution situation of non resource-based cities, but its effect on resource-based cities is not significant. This may be because resource-based cities are cities that develop natural resources such as coal, minerals, and forests, with mining and processing industries as the dominant industries. Although these cities have abundant natural resources, their industrial development relies heavily on resources, and the environmental pollution caused by mining and processing is severe, resulting in the institutional dividends of FTZ establishment for such cities not yet apparent. Compared to resource-based cities, non resource-based cities have a relatively more reasonable and diversified industrial structure, a more complete overall economic structure, and are easier to optimize and upgrade. The gathering capacity of talents, funds, and other factors is also stronger. Therefore, the establishment of FTZ has a more significant effect on improving environmental pollution in non resource-based cities.

The regression results for central and non central cities are shown in columns (3) and (4) of Table 4. The central cities include municipalities directly under the central government, provincial capitals, planned cities, and 35 other cities, including Yangzhou and Xuzhou, totaling 70 large and medium-sized cities in China. Comparing the results of the two columns, it can be found that the regression coefficient of the establishment of FTZ on environmental pollution in central cities is -0.406, which is significant at the 1% level, while the regression coefficient for non central cities is not significant. This indicates that, compared to non central cities, the establishment of FTZ has a more significant impact on the environmental improvement of central cities. This is because the central city has a large population, a wide urban establishment area, complete infrastructure, and a high level of economic development. The improvement effect brought about by the establishment of FTZ is more obvious, and the regional radiation is relatively wider. Environmental pollution is closely related to economic and urban establishment, and the establishment of FTZ will have a more significant environmental impact on the central city.

Mechanism Analysis

The previous tests have shown that the establishment of FTZ can effectively reduce regional pollution emissions, and the results are robust. So what are the possible driving factors behind this effect? This article tests using an intermediary model.

1. The path to upgrading industrial structures. According to column (2) of Table 5, the regression coefficient of the establishment of FTZ on industrial structure upgrading is significantly positive at the 1% level, indicating that the establishment of FTZ has a significant positive promoting effect on urban industrial structure upgrading. According to column (3), the coefficient of industrial structure upgrading is significantly negative, indicating that industrial structure upgrading can improve environmental pollution conditions. Comparing the regression coefficients of columns (1) and (3), it can be found that when the industrial structure upgrading variable is used as a mediator variable, the coefficient of column (3) has increased compared to column (1), indicating that industrial structure upgrading is an effective mechanism for the establishment of FTZ to suppress urban environmental pollution and further indicating that industrial structure upgrading has a positive effect on improving environmental quality.

2. The path of technological innovation. According to column (4) of Table 4, the coefficient of the core explanatory variable FTZ establishment (DID) is 0.537, which is significant at the 1% level, indicating that the establishment of FTZ has a positive promoting effect on the level of technological innovation; that is, the establishment of FTZ has improved the level of urban technological innovation. At the same time, it can be seen from column (5) that the improvement of technological innovation level significantly suppresses the aggravation of urban environmental pollution, and when technological innovation serves as an intermediary path, the inhibitory effect of FTZ establishment on environmental pollution in column (5) has decreased compared to column (1), indicating that technological innovation level has an indirect positive effect on the improvement of environmental pollution in the process of FTZ establishment. This also indicates that technological innovation is an effective intermediary pathway for the establishment of FTZ to improve environmental pollution. From the above test results, it can be seen that industrial structure upgrading and technological innovation are both effective paths for the establishment of FTZ to suppress environmental pollution. Therefore, we have proven hypothesis 1.

Spillover Effect Test

The previous empirical analysis has verified the improvement effect of FTZ establishment on environmental pollution. As the establishment of FTZ not only adds economic growth momentum to the city, but also promotes technology spillover and innovation in adjacent areas, in this section, we examine the spillover effects of FTZ establishment. Specifically, we referred to the method of Zhong et al. (2021) and constructed a spatial weight matrix [15]. Using the spatial Durbin model, we examined the spillover effect of FTZ establishment on urban pollution. From the results in Table 6, we can see that the establishment of FTZ has a significant negative impact on local environmental pollution at the 1% level, indicating that the establishment of FTZ has an improvement effect on urban environmental pollution, which is consistent with the results of benchmark regression. Specifically, from the results in column (2), the coefficient of spatial spillover effect is significantly negative at the 1% level, which means that the establishment of FTZ not only improves local environmental pollution, but also has a spatial spillover effect on the improvement of urban environmental pollution. Therefore, we have proven hypothesis 2.

The spatial lag coefficient in Table 6 is positive and significant at the 1% level, indicating a strong “spatial spillover effect” in the establishment of FTZ. Based on the regression results of SDM, we further estimate the total effect, direct effect, and indirect effect of each variable. From the results in Table 6, it can be seen that for every 1% increase in intensity, the establishment of FTZ will lead to a 0.747% reduction in overall pollution emissions. However, this result will result in a 0.549% reduction in pollution emissions and a 0.198% reduction in pollution emissions from adjacent areas. This indicates that the establishment of FTZ not only reduces local environmental pollution, but also reduces environmental pollution in surrounding areas. From this conclusion, we can see more clearly the positive role of the establishment of FTZ in regional environmental protection.