Quantitative Study on the Synergistic Effect of China’s Plastic Restriction Policy from 2008 to 2025

Abstract

Since 2008, the Chinese government has been intensively implementing policies to control plastic pollution. This study employs text mining and scientific statistical methods to quantitatively analyze 223 policy documents spanning the period from 2008 to 2025. The novelty of this study is associated with the analysis of temporal coherence, content complementarity and subject collaboration in the field of policy aimed at limiting the use of plastic and the development of a universal methodology for the design of complex environmental policies. The results show that policy releases peaked in 2008 and 2020. The average number of policies issued by each province in the eastern region is the highest, with 6.8 items. In terms of policy content, A3, B3, and C3 are the most prominent policy objectives, means and safeguard measures, respectively. The synergy of policy content indicates that market-oriented policies have stronger implementation, for example, the synergy strength between A2 and B2 is 0.7467. The synergy between legislative and enforcement policies is insufficient. For example, the synergy strength between A1 and B1 is only 0.2903. The regional synergy decreases from southwest to northeast. The northeast region lags comprehensively. In terms of time, the similarity of policy texts between 2021 and 2023 remained stable (0.3256–0.3666). The three-dimensional framework offers an approach to policy synergy evaluation. It indicates that the core governance framework of the Chinese Government has strong continuity. This study also has positive value for global environmental protection, such as reducing plastic pollution.

1. Introduction

With the growth of global plastic production and consumption, plastic pollution has become one of the most significant environmental crises, alongside climate change. The 2021 United Nations Environment Programme (UNEP) report points out that the global annual production of plastic waste has exceeded 300 million tons, of which about one-third enters the ocean, soil, and even the food chain due to management failures [1]. There are over 1.6 million square kilometers of floating garbage in the Pacific Ocean, with “microplastics” ranging in diameter from 0.05 to 0.5 cm accounting for 8% of the total mass and 94% of the total amount of marine plastic waste [2]. These microplastics are produced by the physical degradation [3], photochemical degradation [4] or biodegradation [5] of macroscopic plastics [6]. Microplastics can inhibit the activity of soil microorganisms and hinder their growth and reproduction [7,8]. Microplastics can enter organisms through contact [9], respiration [10], food chain enrichment [11] and other pathways [12]. Studies have found that microplastics can induce cellular toxicity and inflammatory responses, which may lead to increased renal toxicity and become a potential pathogenesis for various types of kidney diseases. Microplastics may cause bladder cancer, prostate cancer, kidney cancer, chronic kidney disease and other related diseases [13]. A study has revealed the mechanism by which microplastics induce brain dysfunction and nerve damage by tracking their movement in blood vessels [14]. These studies all indicate that microplastics pose a significant threat to human health.

As one of the world’s largest producers and consumers of plastics [15], China bears significant responsibility in controlling plastic pollution. In 2007, the General Office of the State Council of China issued a notice on restricting the production, sale and use of plastic shopping bags [16]. It pointed out that starting from June 2008, the production, sale and use of ultra-thin plastic bags were prohibited nationwide, and a paid use system for plastic shopping bags was implemented. However, due to regulatory deficiencies and corporate misconduct, the effectiveness of policies has not been fully unleashed [17]. The rapid development of emerging industries such as food delivery and e-commerce logistics has further exacerbated the complexity of plastic pollution control policies. According to data from the National Bureau of Statistics of China, the express delivery volume in China was 507,104.28 million in 2018, and this increased to 83,357,894.3 million in 2020 [18]. In 2020, the China Development and Reform Commission and the Ministry of Ecology and Environment jointly issued the “Opinions on Further Strengthening the Control of Plastic Pollution” [19]. This document sets stricter prohibitions and establishes a systematic governance framework organized by industry and the state. This release marks the deepening of plastic restriction policies. Under this framework, there is intensive interaction between central and local government. Such interaction has given rise to a large number of supporting policies and local innovative practices. Under this governance framework, the use of agricultural plastic film in China has continuously decreased, falling from 2,603,560.59 tons in 2015 to 2,357,943.82 tons in 2021 [18]. China has become the world’s largest and most diverse testing ground for plastic restriction policies.

There has been in-depth research on plastic restriction policies internationally. At the policy design level, Isaac Omondi et al. [20] pointed out, through their study of plastic bag ban policies in 55 African countries, that there is generally a lack of research on policy synergies. Knoblauch et al. [21] found that many countries’ policies focus on the consumption stage and have relatively weak control over the production process. At the level of behavioral intervention, Pamela Yeow [22] pointed out that the UK’s strategy of relying on moral advice has encountered obstacles. Irina Safitri Zen [23] pointed out that the Malaysian government has increased public participation in plastic pollution control through incentive measures. In terms of implementation effectiveness, Jehangir Alishba [24] believes that Pakistan has effectively reduced over three-quarters of plastic bags through policy interventions. Behuria Pritish [25] pointed out that Rwanda has successfully implemented a plastic ban driven by the tourism economy.

Chinese domestic research focuses on the drawbacks of policy mechanisms. Wang Luozhong et al. [26] reveal the gap between regulatory mechanisms in policy texts and implementation through institutional grammar analysis. Xu Jingjie [27] points out the conflict related to the plastic restriction policy between public interest orientation and individual interests from the perspective of comparative interests. Based on the perspective of historical institutionalism, Zhou Yuanyuan [28] identifies path dependence and key nodes in policy changes.

Although China has made some progress in formulating and promoting plastic restriction policies, existing research still has limitations. Firstly, the analysis of dimensions tends to be one-sided. Most of the current achievements focus on the evolution of policy content (such as keyword frequency) or implementation effects (such as recovery rate). This leads to a lack of research on vertical collaboration between central and local governments, cross-departmental horizontal integration, and temporal coherence. Secondly, the methods and tools lag behind. Existing research relies on case analyses and qualitative research, without fully utilizing scientific methods. This makes it challenging to capture the dynamic evolution patterns hidden in large-scale policy texts. Thirdly, there is a disconnect between theory and practice. Although the theory of policy synergy emphasizes the necessity of multidimensional coordination, empirical research has failed to establish an operable quantitative framework. This leads to the core question of how synergy affects governance effectiveness, which remains at the hypothetical level.

This study proposes a quantitative evaluation method for policy synergy effects. This method uses Chinese language models and statistical techniques to analyze China’s plastic restriction policies by looking at how timing, content and publishing organizations work together in three different ways. Therefore, this study reveals the regional collaboration evolution, content correlation and collaborative relationship of China’s plastic policy, aiming to discover potential patterns and policy issues and provide empirical evidence for policy optimization.

The concept of policy synergy was first proposed by the Organization for Economic Cooperation and Development (OECD) [29]. The synergy between policies is the core element that determines whether environmental governance goals can be effectively and efficiently achieved. Policy coordination not only involves vertical coordination between different levels of government (central and local) but also encompasses horizontal integration across policy domains (such as the production, consumption, recycling and research and development of substitutes), as well as the coherence of policy objectives, means and guarantees in the temporal dimension. The lack of a collaborative policy system can easily lead to problems such as conflicting goals, fragmented execution, resource waste and even regulatory arbitrariness. This imbalance ultimately weakens the overall effectiveness of governance. Therefore, the issue of policy coordination between regions has attracted widespread attention from the academic community. The academic community has discussed the current situation of policy synergy [30], factors influencing policy synergy [31] and policy synergy mechanisms [32].

The main contributions of this study include the following:

Unlike earlier studies that only looked at how policies work together in terms of their content, this study creates a framework that examines the collaboration between policy content, the organizations involved and the timing, offering a fresh viewpoint for better understanding how plastic restriction policies work together.

This study utilizes policy texts to conduct a comprehensive process analysis, ranging from semantic mining to statistical methods.

This study reveals the causal chain that links policy content to governance effectiveness, thereby offering a universal methodology for the design of complex environmental policies.

In summary, this study aims to address the following:

RQ1: How can we build a quantifiable and verifiable evaluation method for the synergistic effects of plastic restriction policies?

RQ2: How does multidimensional synergy (temporal coherence/content complementarity/subject collaboration) present differentiated spatial patterns and dynamic trajectories with policy evolution?

The remainder of this paper is structured as follows: Section 2 develops the method used in the paper. Section 3 presents and discusses the results. Section 4 provides policy recommendations based on the research results. Section 5 sets out the conclusions.

2. Method

This article compiles a total of 223 policy documents related to plastic pollution prevention and control issued by the Chinese central government and provincial-level local governments, excluding Hong Kong, Macao, and Taiwan, since 2008 (40 central documents and 183 local documents). Due to differences in legal systems, policy implementation, and data statistics between Hong Kong, the Macao Special Administrative Region, and Taiwan, these regions were not included in the analysis framework. Using text mining and scientific statistical methods, text analysis was conducted on policy content, region and time.

This study divides the content of policy documents into three dimensions: A, policy purpose; B, policy means; and C, policy guarantee. Each dimension is divided into three sub-dimensions, as defined in

Table 1. Three-dimensional collaborative analysis framework for China’s plastic restriction policy.

Main Dimension	Sub Dimensions and Definitions
A. Policy purpose	A1 (Legislative Restriction): Propose legislative or administrative measures to directly restrict, prohibit or phase out disposable plastic products, including the establishment of negative lists, prohibition catalogs and production and sales control measures.
	A2 (industrial substitution): Emphasis is placed on achieving plastic reduction goals through the use of new environmentally friendly materials, biodegradable packaging, green supply chains and other means, with a focus on market guidance and industrial substitution pathways.
	A3 (Recycling and processing): Focusing on the classification and recycling system of waste plastics, the capacity for recycling and processing and the construction and operation of garbage removal and harmless treatment facilities.
B. Policy means	B1 (laws and regulations): Involving the formulation or revision of laws and regulations, policy documents, management measures, technical standards, directories, licensing systems and other regulatory measures.
	B2 (Economic incentives): Involving economic incentive measures such as fiscal subsidies, tax incentives, green finance, special funds and price regulation.
	B3 (science and technology): Involving technology research and development, achievement transformation, scientific research and development, platform construction, scientific and technological support, etc.
C. Policy guarantee	C1 (Multi departmental collaboration): Involving the establishment of a multi-departmental linkage mechanism, task division, responsibility assessment, supervision, execution and other organizational support systems.
	C2 (administrative supervision): Involving law enforcement inspections, daily supervision, investigations and punishment of violations; credit penalties; and other related matters.
	C3 (publicity and education): Carrying out non-coercive measures such as public education, media publicity, social advocacy and promotion of typical cases.

. This study conducted the following analysis: Firstly, we used the Chinese semantic model text2vec-base-chinese to vectorize all files, calculated cosine similarity using Equation (1), and counted the frequency of occurrence of 9 dimensions, including A, B, C, and their sub-dimensions. Secondly, we counted the number of files that appear together in any two of the nine dimensions and obtained the co-occurrence matrix based on Equation (2). Thirdly, we calculated the degree of synergy between any two dimensions based on the co-occurrence situation and Equation (3).

This study conducted an analysis of regional synergy by calculating the cosine similarity (1) between provincial policy texts and central policy texts. The cosine similarity value ranged from 0 to 1. The closer the number was to 1, the more relevant the text; the closer it was to 0, the more irrelevant the text.

In addition, this study specifically selected the period from 2020 to 2024, which had the highest number of documents issued, to analyze the synergy of policy content over the past five years. Firstly, we conducted keyword statistics analysis on policy documents, selected the 30 most frequently appearing keywords related to policy content, and used the TF-IDF method involved in Equations (4)–(6) to calculate the average distribution of these keywords in each year’s documents.

The specific calculation method is as follows:

$$Sim(A,B)=\cos (\theta )={\displaystyle \frac{A·B}{‖A‖·‖B‖}}$$

(1)

In the formula, A and B are vector representations of paragraph and dimension descriptions. A·B is the dot product, $‖A‖$ and $‖B‖$ are the magnitudes of vectors.

$$C_{y,z}={\displaystyle \sum _{i=1}^N\mathbb{I}\left(y\in D_i\wedge z\in D_i\right)}$$

(2)

In the formula, N is the total number of documents, $D_i$ is all the policy dimensions identified in the document, and $\mathbb{I}$ is the indicator function (returns 1 if the condition is met; otherwise, it returns 0).

$$S_{\left(y,z\right)}={\displaystyle \frac{f_{yz}}{\min (f_y,f_z)}}$$

(3)

In the formula, $S_{(y,z)}$ is the synergistic effect value of dimensions y and $f_{yz}$ is calculated as the number of documents containing both dimensions y and z (co-occurrence frequency), where $f_y$ is the total number of documents containing dimension y and $f_z$ is the total number of documents containing dimension z.

$$TF(k,d)={\displaystyle \frac{count(k in d)}{{\displaystyle {\sum }_{w\in d}count(w)}}}$$

(4)

$$IDF(k)=\log \left({\displaystyle \frac{N}{\left|\left\{d\in D:k\in d\right\}\right|+1}}\right)$$

(5)

$$TF-IDF\left(k,d\right)=TF\left(k,d\right)\times IDF\left(k\right)$$

(6)

$$AvgTF-IDF\left(k,y\right)={\displaystyle \frac{1}{\left|D_y\right|}}{\displaystyle \sum _{d\in D_y}TF-IDF\left(k,d\right)}$$

(7)

Equation (4) represents the frequency of keyword k appearing in document d, normalized by the total number of words in the document. N represents the total number of documents, and $\left|\left\{d\in D:k\in d\right\}\right|$ represents the number of documents containing the keyword k. Equation (7) is used to calculate the average TF-IDF value for all documents within year y.

This article uses Python stable version 3.13 to call the text2vec base Chinese semantic model for training. Text2vec base Chinese is a Chinese sentence embedding model based on the BERT architecture that has strong contextual understanding ability and can capture the meaning changes of words in different contexts. This is particularly important for policy texts, which usually contain complex structures and professional terminology. Specifically, by scientifically evaluating the synergistic effects between different plastic restriction policies, decision-makers can identify the optimal combination of policies; avoid conflicts or resource waste in terms of time, space and other dimensions; and significantly improve overall governance efficiency. For example, when there is positive synergy between the “A1 dimension” and “B1 dimension” policies, their joint pollution reduction effect may be much greater than the sum of the individual policies. On the contrary, if there is negative synergy between policies, it may lead to the failure of governance goals. This systematic evaluation is a crucial step in promoting plastic pollution control from a “single policy” to a “synergistic effect” and is of great significance for achieving sustainable development goals.

3. Results

3.1. Plastic Restriction Policy Spatiotemporal Evolution Analysis

The theory of regional policy synergy reveals that breaking down administrative barriers through institutional innovation and integrating fragmented local actions into a cohesive governance force can effectively address the externalities of public affairs. Figure 1 shows the time distribution of China’s plastic restriction policy since 2008. According to the sample selection criteria, a total of 40 policy documents related to plastic restrictions were issued at the national level between 2008 and 2025. Among them, there were a large number of policies issued in 2008 and 2020, with 6 and 16, respectively. In 2008, the General Office of the State Council of China issued a notice restricting the production, sale and use of plastic shopping bags. This document proposes the paid use of plastic shopping bags in supermarkets, shopping malls and other places. It prohibits the production, sale and use of ultra-thin plastic shopping bags with a thickness of less than 0.025 mm. This document marks the official launch of the plastic restriction work. In 2020, a series of policies focused on the “Opinions on Further Strengthening the Control of Plastic Pollution” were released in succession. These policies propose to restrict non-degradable plastic products in stages, doing so by region and by category. This process means that plastic pollution control is becoming more mature.

The number of local government documents is similar to that of the central government. After the release of seven policies by provincial governments in 2008, no more were introduced in a centralized manner. By 2020, in response to central policies, provincial governments had issued a total of 53 policy documents. Over the next five years, the number of policies issued by provincial governments decreased annually. It is worth noting that there has been no decrease in the number of policies in 2023 compared to the previous year. This is because the central government issued eight new policies in 2023, including the Notice on Solid Promotion of Plastic Pollution Control, in response to the special stage of the COVID-19 epidemic. Provincial governments have also made corresponding adjustments.

Figure 2 shows the regional distribution of plastic restriction policies in China. According to the Several Opinions of the Central Committee of the Communist Party of China and the State Council on Promoting the Rise of the Central Region, the economic regions have been divided. This article categorizes provincial-level administrative units in China, excluding Hong Kong, Macau, and Taiwan, into four regions: eastern, western, central, and northeastern. This article summarizes the cumulative number of policies issued by various provincial units, the proportion of policies issued by each region, and the average number of policies issued by provincial units in each region. There is a significant difference in the number of plastic restriction policies issued between 2008 and 2025. The eastern region has an average of 6.8 policies introduced by each province. For the western region, the number of policies introduced by each province is the lowest at only 5.3. The average number of policies released in the central and northeastern regions is 6.2 and 5.7, respectively. Regional development characteristics and policy demands mainly drive this distribution pattern.

As shown in

Table 2. Gross products of various regions in China in 2020 (unit: billion yuan) [33].

Region	Province	GDP
Western Region (Average Value: 18,079.6)	Chongqing City	25,158.1
	Sichuan Province	49,445.1
	Yunnan Province	25,214.5
	Guizhou Province	18,308.3
	Guangxi Zhuang Autonomous Region	22,250.7
	Xizang Autonomous Region	1956.5
	Shaanxi province	26,297.0
	Gansu Province	9323.1
	Ningxia Hui Autonomous Region	4036.2
	Qinghai Province	3080.6
	Xinjiang Uygur Autonomous Region	14,262.2
	Inner Mongolia Autonomous Region	17,623.4
Eastern Region (The Average Value: 54,027.9)	Beijing City	38,503.6
	Tianjin City	14,230.8
	Hebei Province	36,821.5
	Shanghai City	41,603.9
	Jiangsu Province	104,566.6
	Zhejiang Province	67,164.5
	Fujian Province	43,682.0
	Shandog Province	74,355.9
	Guangdong Province	113,708.9
	Hainan Province	5640.8
Central Region (Average Value: 36,921.5)	Shanxi Province	18,202.7
	Henan Province	54,160.6
	Anhui Province	38,628.8
	Hubei Province	43,017.6
	Jiangxi Province	25,825.4
	Hunan Province	41,693.7
Northeastern Region (Average Value: 17,446.2)	Liaoning Province	25,839.0
	Jilin Province	12,499.5
	Heilongjiang Province	14,000.1

and Figure 2, the higher the GDP of a region, the more plastic restriction policies are issued on average. This indicates that GDP is directly proportional to the number of plastic restriction policies issued. High-GDP regions are usually accompanied by higher per capita consumption levels (such as express delivery, takeaway, retail packaging, etc.), directly driving up the amount of plastic waste. The production of plastic packaging, building materials and daily necessities is concentrated in industrial and service industry clusters such as the Yangtze River Delta and the Pearl River Delta. Their demand for source control is stronger. As shown in Table 2, the average GDP of the eastern region in 2020 was 5402.79 billion yuan, which is the highest among the four regions. The average number of plastic restriction policies issued in the eastern region is also the highest. The GDP of the northeastern region is the lowest, and the average number of plastic restriction policies issued is also the lowest. On one hand, this is due to the high demand for solving plastic pollution in the eastern region, and on the other hand, the eastern region has sufficient financial capacity to support the treatment of plastic pollution. For example, in 2020, the local fiscal expenditure on environmental protection in Jiangsu Province reached 33.69 billion yuan, while in Liaoning Province, this expenditure was only 9.794 billion yuan [33].

The policy quantity in the western region ranks third and is directly related to the ecological function of the western region. As the source of the Yangtze and Yellow Rivers and an ecologically fragile area, plastic pollution poses a serious threat to soil and water conservation. The pollution of plastic film in the arid northwest region has led to soil compaction and reduced agricultural yield [34]. The country allocated resources through the Western Development strategy, and from 2008 to 2016, the central special fiscal transfer payment promoted local policies such as agricultural film control [35]. Xinjiang, Xizang and other regions demonstrated their commitment to ecological governance through the implementation of a plastic ban policy.

This difference is a microcosm of uneven regional development in China. The eastern region and central region lead governance innovation with economic advantages, while the western region utilizes national strategies to protect ecological barriers. The northeastern region, in contrast, seeks a gradual path to transformation. Compared to the past studies where plastic restrictions were only sorted by time [36], a regional comparison of policies helps to link policies with their deep backgrounds such as local economies, industries, resources, and development stages. The significance of regional comparison lies in revealing the differences in policy response, scope, intensity and supporting measures among different regions. This can analyze the economic and geographical factors behind the differences. It is useful to lay the foundation for optimizing policies and evaluating their effects according to local conditions, as well as to promote the more effective and sustainable implementation of plastic limit regulations.

3.2. Collaborative Analysis of Policy Content

The collaborative theory emphasizes the collective collaboration process of various subsystems under the guidance of scientific policies, relying on collaborative networks constructed by different subsystems to achieve an overall effect of 1 + 1 > 2. Figure 3 shows the statistical results of the policy elements of China’s plastic restriction policy. In terms of policy purpose, A3 is most valued (97 times), followed by A1 (88 times), and A2 has the lowest frequency (75 times). A3 prioritizes the classification and recycling of waste plastics as its policy core, ensuring a practical and cost-effective approach. Plastic pollution can be directly reduced through garbage classification and resource utilization, without the need to change existing production and consumption patterns. A1 is centered around prohibition. A1 is the leading force in the early stages of plastic pollution control. As time goes by, the effectiveness of A1 gradually decreases. A2 focuses on green substitution, and its promotion is constrained by technology and cost.

In terms of policy measures, B3 has the highest frequency (116 items), followed by B2 (110 items), and B1 has the lowest frequency (62 items). B3 focuses on technology research and development. The production of biodegradable plastics has faced a serious problem of insufficient production capacity. Government policies promote industrial upgrading through research and development subsidies and technical standards to achieve the goal of material substitution. B2 focuses on economic incentives as its core. The government adopts procurement subsidies and tax reduction policies to balance the obstacles to enterprise transformation. B1 focuses on formulating laws and regulations, mainly limited by the difficulty of implementation and social risks.

In terms of policy protection, C3 has the highest frequency (109 items), followed by C2 (105 items), and C1 has the lowest frequency (81 items). C1 is centered on multi-departmental collaboration, involving multiple departments such as the National Development and Reform Commission, the Market Supervision Bureau and the Commerce Department, making it challenging to achieve practical multi-departmental cooperation. C2 focuses on law enforcement and supervision. However, because of high enforcement costs, covert violations and widespread misconduct, law enforcement is a relatively complex issue. C3 focuses on publicity and education. The public readily accepts C3, and it has become a low-cost priority option.

The distribution of policy tools reflects the path of China’s environmental governance. China’s environmental governance is led by technological tools, coordinated by economic incentives and supported by legal means.

Figure 4 illustrates the synergistic effects among all sub-dimensions of policy content, which demonstrates the multifaceted and inter-related nature of policy issues. The triangular matrix heatmap represents the collaborative strength between two dimensions, ranging from 0 to 1. The higher the value, the greater the synergy between sub-dimensions.

The internal correlations among dimension A, dimension B, and dimension C are all strong. The synergy between A2 and A3 reaches 0.76. This indicates that policy design has formed highly collaborative sub-dimensions within the same dimension. For example, in dimension A, the increase in plastic pollution is first reduced through A1. The supply structure is transformed through A2, and finally, the stock is processed through the A3 recycling system. The B dimension is a complementary design. B1 provides coercive force, B2 provides economic power and B3 provides solutions. The C dimension is a progressive relationship of execution. The C dimension forms a mechanism of C1 accountability, C2 correction and C3 prevention.

In the A and C dimensions, the correlation between A3 and C3 is 0.5773. This indicates that social participation policies rely on flexible guidance. The effectiveness of the A3 dimension depends on civic awareness. For example, garbage classification requires active cooperation from the public. C3 can reduce behavioral resistance through advertising. A3 and C3 have obvious advantages in terms of low cost and wide coverage. Compared to C2, C3 can penetrate scenarios such as homes and campuses. The synergy between A1 and C1 is only 0.3704, which indicates that there is insufficient support for achieving the goal. When market regulatory authorities investigate and deal with the circulation process, they have no right to shut down production workshops. The environmental protection department is responsible for terminal pollution control and has no authority to punish merchants. The National Development and Reform Commission is responsible for overall planning and coordination but lacks a drive for local assessment.

In the A and B dimensions, the synergy strength between A2 and B2 is 0.7467. Market-oriented policies are easier to implement. This effect is driven by cost resistance. Due to the lack of motivation for enterprises to use high-priced, environmentally friendly materials, B2 directly offsets costs through subsidies, thereby breaking through the implementation bottleneck of A2. In addition, policy design conforms to economic laws, guiding industrial transformation through incentives rather than coercion and forming a sustainable business closed loop. The synergy strength between A1 and B1 is only 0.2903, indicating a dual failure of legislation and law enforcement.

In the B and C dimensions, the synergy between B3 and C1 is 0.6296. This indicates that technological innovation requires institutional support. Technical breakthroughs require cross-departmental coordination of resources and expertise. Departmental barriers can be broken down by the linkage mechanism of C1. This feature also reflects long-term support. Long-term technology research and development cycles necessitate C1’s assessment and supervision to maintain policy continuity. The synergy between B1 and C1 is 0.4516. This value indicates that institutional deficiencies have weakened the legal effect. Although implementation relies on C1’s supervision and assessment, there is a lack of normalized mechanisms for grassroots law enforcement. The credit punishment measures of C2 are challenging to implement across departments, and their legal effectiveness is weakened.

Through a three-dimensional collaborative analysis of policy purpose, policy means and policy guarantees, this study goes beyond the simple statistical analysis of one dimension conducted in previous research [37], revealing the reasons for the scarcity of plastic restriction order punishment rules (B1). China’s plastic restriction policy is centered around B3 and B2, using technology and subsidies to replace mandatory punishment and form an efficient path. Specifically, the high-frequency technology promotion combination (such as A3-B3-C3) avoids high law enforcement costs. B2 effectively breaks the cost resistance of A2 and reduces dependence on B1. This is consistent with what some scholars have pointed out, stating that economic policy instruments, especially economic incentives, have recently been gaining popularity [38]. A1 and B1 have low synergy (0.29) due to difficult execution, indicating that rigid methods are inefficient.

3.3. Policy Issuer Analysis

Policy entities should support and closely cooperate with each other, forming a strong policy synergy to effectively solve complex cross-regional problems and jointly achieve policy goals. Figure 5, Figure 6 and Figure 7 measure how well the plastic pollution prevention and control policies from 31 provincial-level administrative units in China match with those from the central government in three areas of text content: A, B and C. They are presented in the form of cosine similarity. This indicator reflects the degree of overlap between local policies and central policies. The higher the value, the closer the local policies are to the overall strategy of the central government.

In terms of overall synergy, the average synergy of policy purpose A is the highest at 0.788, which is due to the rigid transmission of goals led by the central government. The average synergy for policy guarantee dimension C is 0.741. This value reflects the limited adaptability of the institutional framework. Legal and organizational protection needs to rely on the existing administrative system. Thus, it should adjust organizational configuration based on financial capacity. The average synergy of policy tool dimension B is the lowest at 0.661. This characteristic stems from its direct linkage to local resources, resulting in significant differences between the eastern and western regions. The industrial foundation severely constrains the use of technological tools. Economic incentives rely on local financial resources. The shortage of grassroots law enforcement personnel in the legal field presents a significant differentiation. This forces underdeveloped areas to weaken punishment and shift toward a propaganda orientation.

In terms of regional synergy, the northeastern region lags comprehensively. Both the B dimension (0.630) and the C dimension (0.715) in this region are the lowest in the country. This is due to the problems of industrial decline [39] and population outflow [40] in the transformation momentum of the northeastern region. This scenario has led to shortcomings in policy implementation and guarantee mechanisms in Northeast China. The synergy of various dimensions is most prominent in the western region. This indicates that the western region can effectively utilize policy tools to implement central goals. Among them, the key driving force is the precise injection of central resources. Many provinces in the western region have fragile and important ecosystems. Their importance requires stricter measures to be taken for protection. The average policy synergy in the eastern region is generally low, especially in economically developed areas such as Jiangsu, Beijing and Shanghai. For example, Jiangsu ranks only 29th in terms of A-dimensional synergy. Shanghai ranks 27th in terms of B-dimensional collaboration. Beijing ranks 30th in terms of C-dimensional collaboration. This ranking reflects the unique demand for plastic pollution control in developed regions, for example, the catering and express delivery industries. The synergy performance in the central region is balanced. The synergies of all dimensions are in the middle range (A: 0.791, B: 0.654, C: 0.744), and there are no apparent weaknesses or advantages.

Figure 8 shows the overall differences in text content synergy between provincial policies and national policies. The primary manifestation of regional differences is the gradual decrease in synergy from southwestern to northeastern regions. Among them, Xizang, Xinjiang, Tianjin, Hebei, Ningxia and Chongqing demonstrate high synergy. Their synergies are greater than 0.74. This value reflects the special governance mechanism of border areas and the administrative efficiency advantages of municipalities directly under the central government. These regions often ensure the precise implementation of national directives through policies. The synergies of Guangdong, Jiangxi, Anhui, Hubei, Shanxi, Jilin, Liaoning and other places rank second. They are between 0.72 and 0.74. This ratio reflects the mainstream execution paradigm. This range also reflects the universality of policies in conventional economic regions. In contrast, the synergistic effects in Heilongjiang, Inner Mongolia, Shanxi, Henan, Shandong, Zhejiang, Shanghai and Jiangsu are relatively low. Their values are all below 0.7.

The reasons for this are also not entirely the same. Shanxi is a resource-rich province; for example, Shanxi has abundant coal resources. It is facing the dilemma of transferring industry substitution costs associated with the plastic industry. Jiangsu, Zhejiang and other provinces have strong manufacturing industries. They are also major producers, consumers and exporters of plastics. The complexity of their industrial chains has led to a decline in policy transmission. Henan, Shandong and other provinces are major agricultural regions in the country. The agricultural plastic film is a significant source of microplastics in the soil of farmland throughout China. Plastic bundling film and feed packaging are widely used in Inner Mongolia grassland pastoral areas. Due to the vast territory and sparse population of Inner Mongolia, the cost of recycling is exceptionally high.

3.4. Analysis of Policy Time Synergy

Time synergy emphasizes the connection and continuity of new and old policies in terms of content. As shown in Figure 9, we selected the top 30 effective keywords with the highest frequency of occurrence from 223 policy documents and calculated their proportion in the policy documents from 2020 to 2024. The width of a keyword represents its TF-IDF value in the current policy. The continuous change in width reflects the shift in strategic focus.

The weight of the keyword ‘Plastic’ has consistently remained high but has decreased over the years. The weight of ‘Plastic’ decreased from 0.615 in 2020 to 0.271 in 2024. This word is usually used in conjunction with mandatory measures such as “prohibition”, “reduction” and “restriction”. The decrease in its proportion marks a gradual shift in policy focus from initial, relatively singular consumer-end “ban and restriction” control to more systematic governance. This change reflects policymakers’ recognition that relying solely on end-of-pipe restrictions makes it difficult to cure the problem of plastic pollution.

In contrast to the decline in “Plastic”, there has been a significant increase in the proportion of “Green”. The growth of “Green” is accompanied by the synchronous growth of “Cleaning” and “Recycling”. This phenomenon expresses a shift in policy focus. The rise of “Green” represents a shift in development philosophy. It emphasizes the environmentally friendly properties of plastic products and the greening of the entire industry chain. The strengthening of “Cleaning” and “Recycling” focuses on the collection, classification, resource utilization and harmless disposal of waste generated. This data indicates that China is vigorously building and improving its plastic waste recycling system as a key step in addressing pollution and resource waste.

The weights of “Packaging” and “Express Delivery” are high, and their changing trends are highly consistent. This indicates that the plastic packaging pollution caused by emerging industries such as express delivery and food delivery has been placed at the core of the policy agenda. Research has shown that 2020 was another critical year in China’s plastic regulation history, marking the beginning of a new stage and approach concentrating explicitly on the governance of specific plastic types such as disposable plastic products, express packaging and fertilizer packages, as well as specific stages of the plastic life cycle, including the use, collection, recycling and reuse of various plastics [38]. According to data from the National Bureau of Statistics of China, the volume of express delivery in China increased from 833,578,943 million in 2020 to 1,750,838,583 million in 2024 [18]. China is focusing on addressing this source of pollution and promoting reductions in packaging, standardization and recycling.

The TF-IDF of “Disposable”, “Takeout” and “Catering” decreases synchronously. This indicates that the governance of disposable plastic consumer goods has entered a new stage. According to the 46th Statistical Report on the State of China’s Internet in 2020, during the COVID-19 pandemic in the first half of 2020, some catering stores and residential areas were closed, reducing users’ demand for takeout [41]. In 2020, with the release of the “Opinions on Further Strengthening Plastic Pollution Control”, the popularity of policy keywords naturally declined. This trend indicates that a basic control framework has been established and has entered the stage of normalized supervision. The weights of “Packaging” and “Express Delivery” significantly increased during the same period. Such an increase indicates that policy resources are skewed towards the e-commerce express delivery sector, which has experienced faster growth and more pronounced pollution increases.

The overall increase in “Agriculture” is an important signal. This primarily highlights the increasingly severe issue of agricultural film residue pollution and its elevated status from a policy perspective. Plastic pollution in the agricultural sector is dispersed and difficult to recycle. The increasing proportion of agriculture indicates that China is increasing its attention to and governance efforts on this issue.

The “Industry” and “Energy Conservation” sectors continue to grow and exhibit a consistent trend. This reveals that policies are closer to the source of plastic pollution—the production and manufacturing process. The dynamic changes in the weights of these keywords reflect the development trajectory of China’s plastic restriction policy, ranging from prohibitions and restrictions on the consumer end to the establishment of a waste recycling system. The policies will further deepen their focus on green substitution at source, industrial transformation and upgrading and complete lifecycle management. Policymakers are increasingly inclined to adopt systematic and comprehensive solutions. They will bring together methods like cleaning up waste at the end, improving recycling processes, controlling pollution at source with green production and focusing on important areas like packaging and agricultural film to better tackle the complicated issues of plastic pollution. Their aim is also to serve the overall strategy of “dual carbon” goals and ecological civilization construction.

Figure 10 quantifies the temporal collaborative evolution characteristics of China’s plastic restriction policy theme based on cosine similarity. The blue line represents the cross-period between the policies of each year and 2020 as the benchmark year. The implementation opinions on further strengthening plastic pollution control were issued by each province in 2020. The similarity remained stable in the range of 0.3256–0.3666 from 2021 to 2023. This pattern indicates that the core framework for plastic pollution control has strong continuity. The trend reflects the close adherence of provincial policies to central documents, ensuring the continuity of governance. In 2024, the similarity dropped to 0.2427. The decrease marks a shift in policy direction. The variation is consistent with the decrease in the proportion of “Plastic” and the increases in “Green” and “Recycling” in the aforementioned keyword analysis. This indicates that the policy focus has shifted from the initial “prohibition and control” strategy to “circular substitution”. For instance, policies are bolstering the research and development of biodegradable materials and recycling infrastructure.

The red line represents the periodic fluctuations of dynamic collaboration between adjacent years. Compared with the policy in 2021, the synergy in 2022 was 0.3643. Compared to 2022, the policy synergy in 2023 decreased to 0.3197. Compared to 2020, the policy synergy in 2023 was expected to reach 0.366. This number indicates a policy correction from 2023 to 2020. The result is an improvement in the 2020 Implementation Opinions on Further Strengthening Plastic Pollution Control. The synergy of policies in 2024 compared to 2023 is 0.3422. This value indicates that the differences in local pilot programs have been partially integrated. Compared to 2020, the synergy in 2024 is only 0.2427. This figure suggests that the policy system has become bifurcated, with short-term operational adjustments and long-term strategic transformations.

4. Policy Recommendations

In terms of policy content, since A1 has the weakest connection to B1, it is advisable to turn the main achievements of A1 (like negative lists and prohibition catalogs) into rules or required standards for B1. This can provide a legal basis for the effective and unified execution of direct control measures, thereby enhancing policy synergy and operational effectiveness. To address the finding of a weak connection between A3 and B1, it is suggested that the important technical requirements of A3 (like classification standards, recycled material quality control and processing facility specifications) be turned into required technical standards and operating rules for B1. These changes can provide a legal basis and unified regulatory standards for the efficient operation of the recycling system. This enhances policy synergy and implementation efficiency. To improve cooperation between A1 and C1, it is advisable to include the main goals and specific rules of A1 measures (like negative lists and sales bans) in the joint action plans and yearly performance assessments of related departments (like ecological environment, market supervision, commerce, etc.) (C1). These steps can strengthen departmental execution motivation and collaborative accountability. Such an approach can enhance the effectiveness of regulatory measures and policy implementation.

In terms of geography, to address the issue of the lowest central coordination in the B dimension, policymakers at the national level should systematically sort out and integrate B-class policy measures (such as standards, permits and directives) that involve production, circulation, consumption and recycling; develop unified guidelines for the collaborative application of plastic restriction policy tools; and clarify the scope, priority and collaboration rules of various tools to reduce ambiguity and conflicts in local execution. In response to the pattern of “high synergy between western and southern regions, and low synergy between eastern and northern regions”, it is recommended to implement classified guidance. For economically developed regions with strong governance capabilities, such as the eastern and northern regions, we will focus on strengthening policy innovation and the construction of cross-regional coordination mechanisms. It is advisable to encourage the exploration of higher standard collaborative models within the national framework, such as joint supervision of metropolitan areas and unified technology platforms. Efforts will be made to consolidate the existing collaborative achievements in regions with good collaborative foundations, such as the western and southern regions; strengthen capacity building and resource guarantee; promote the upgrading of the collaborative mode from “passive response” to “active optimization”; and establish a nationwide dynamic monitoring and evaluation mechanism for the synergy coefficient of plastic restriction policies, providing data support for precise policy implementation.

In response to the lack of a long-term mechanism in the time dimension and the lag in policy focus adjustment, it is recommended to establish a mandatory policy evaluation cycle every 3–5 years. Based on environmental performance, social feedback and technological progress, policymakers should systematically revise regulatory standards (B1) and control lists (A1/A3) to ensure the continued effectiveness of policies. Based on annual keyword trend analysis (such as “green recycling”, “recycling” and “alternative materials”), they should also establish special funds and research and development plans and guide policy resources to tilt towards emerging key areas such as biodegradation technology, efficient recycling models and green packaging innovation, as well as accelerate technological breakthroughs and application promotion.

5. Conclusions

This study uses text mining and statistical methods to conduct text analysis on a total of 223 plastic restriction policies issued by the central government and provincial governments, except for Hong Kong, Macao and Taiwan, from 2008 to 2025. The following conclusions can be drawn from three aspects: policy content, region and time.

In terms of time distribution, the two peak periods for the release of China’s plastic restriction policy were 2008 (6 items from the central government/7 items from local governments) and 2020 (16 items from the central government/53 items from local governments). There are significant differences among provinces. The average number of policies issued by western provinces is the lowest (5.3 items). The eastern provinces have the highest average number of policies released (6.8). This distribution reflects the imbalance in regional development. The eastern region has achieved refined control through high-frequency policies. The western region is guided by ecological priority. The northeastern region is facing the problem of insufficient policy supply due to lagging transformation.

In terms of policy coordination, China’s plastic restriction policy follows a technology-led and market-driven approach. B3, with technology research and development as its core, has the highest frequency (appearing 116 times). This indicates that solving the substitution problem involves industrial upgrading. B2, with economic incentives as its core, appears 110 times. B2 has the highest synergy with A2, which is centered around green substitution (0.75). This indicates that the subsidy policy effectively reduces the cost of transformation. B1, which focuses on formulating laws and regulations, has only been used 62 times. This situation reflects the execution dilemma (A1–B1 synergy degree = 0.29). The weak C1 centered on cross-departmental collaboration has led to a regulatory gap (A1–C1 synergy degree = 0.37).

In terms of regional coordination, China’s plastic restriction policy presents a collaborative feature of “unified goals and differentiated execution”. The central and local governments exhibit a high degree of consistency in their policy objectives, as indicated by the national average of 0.7881 in the A dimension. The synergy degree of policy measures (B-dimension) is the lowest (0.66). This statistic highlights the impact of differences in resource endowments. The northeastern region lags behind severely due to industrial decline (with a B-dimension of only 0.63). Developed eastern provinces (Jiangsu/Shanghai) require policy innovation space due to complex industrial chains. The western frontier region (Xizang/Xinjiang > 0.74) implements central directives through an efficient implementation mechanism. The interprovincial synergy decreases from southwest to northeast.

In terms of policy time coordination, keyword analysis of China’s plastic restriction policy reveals that the policy focus has shifted from end-of-pipe control (with a weight of 0.615 for “plastic” in 2020) to systematic governance. In 2024, both “green” and “recycling” increased, and “agriculture” continued to strengthen. This forms a three-in-one model of “source substitution process recycling key focus”. Collaborative data confirms the characteristics of transformation. The cross-period synergy between 2021 and 2023 remains stable in the range of 0.3256–0.3666. The range reflects the continuity of the policy framework. It dropped sharply to 0.2427 in 2024, marking the upgrade of the system. This decrease provides mutual evidence, with the weight of “plastic” being reduced to 0.27055. Adjacent collaborative fluctuations reveal new mechanisms of interaction between central and local government. The low point in 2023 (0.3197) corresponds to the expansion of local pilot projects. The 2024 pullback (0.3422) reflects the integration of innovative experience.

In terms of policy recommendations, firstly, institutional transformation should be carried out, embedding A1/A3 core control requirements into regulatory standards (B1) and departmental assessments (C1). The second step is differentiated collaboration, building a central guidance framework and regional classification governance model. The third step is dynamic iteration, establishing a periodic evaluation mechanism and a policy response system empowered by technology. By strengthening three-dimensional collaboration, we aim to enhance the legal effectiveness, enforcement rigidity and sustained innovation capability of the policy system.

While this study makes important theoretical and practical contributions to the spatiotemporal synergy, content synergy and publishing subject synergy of China’s plastic restriction policy texts from 2008 to 2025, several limitations warrant acknowledgment and present opportunities for future research.

There is currently a lack of unified quantitative norms in the text on plastic restriction policies regarding production standards, circulation supervision, definition of consumption scenarios (such as whether takeaway plastic is disposable) and requirements for the recycling and disposal of plastic products. This makes it difficult to effectively perform cross-regional comparisons and leads to difficulties in monitoring policy effectiveness. However, the “Opinions on Further Strengthening the Control of Plastic Pollution” clarifies the goal of establishing a management system for the production, circulation, consumption and recycling of plastic products by 2025. This marks the institutional transition of plastic restriction governance from fragmented control to standardized management throughout the entire lifecycle. Future research can utilize these standardized indicators to deepen the evaluation of cross-regional synergy, causal analysis of policy effectiveness and precise measurement of policy evolution based on time series.

Although our current analysis focuses on the collaborative characteristics of policy time, content and publishing subjects, revealing their collaborative relationships, subsequent research can expand the multidimensional interpretation of policy texts by introducing perspectives such as policy implementation effectiveness evaluation, public opinion and social response analysis and cross-border policy comparison, combined with intelligent text mining methods such as machine learning. Exploring these directions will not only help to fill the gaps in current research on policy implementation effectiveness and external influences but also deepen the systematic understanding of the connotation and evolutionary mechanism of plastic restriction policy texts.