Introduction

In the past half century, the diffusion of policy innovation has been a major research topic in academia, especially in research on American politics and public policy [3, 5, 21]. The study of diffusion of policy innovation as an objective political phenomenon or as a theory of policy process has been improved in recent decades, especially in the twenty-first century, and it has accumulated significant research results [18]. However, existing research mainly focuses on the innovation and policy diffusion of policy inventors and pays little attention to the innovations by adopters during policy diffusion [54, 64, 72].

Traditional research contends that policy diffusion will gradually lead to similarity in policies adopted by different regions [16]. However, in the policy diffusion process, scholars have found that policy innovation adopters often engage in policy reinvention using knowledge gained from other governments. Scholars have called on academia to pay attention to this phenomenon. Policy reinvention has become a new direction in policy process research [8, 19, 44, 47].

Most of the existing policy reinvention research focuses on political systems that feature federalism and electoral democracy (mainly in the United States). At present, many political and public policy scholars have realized this research should not focus exclusively on Western political systems [2, 65, 67, 76]. Commonly accepted theoretical hypotheses developed for Western countries must be tested empirically in diverse contexts, such as in developing countries [40]. In contrast to the federal systems (the separation of powers between the central government and the local government) and democracy (the official election system), the Chinese government is more decentralized in economic and social responsibilities and centralized in organizational and personnel systems. The central government has full authority to intervene in the actions of provincial governments and uses performance evaluation instead of constituent election to select provincial officials [76, 78]. Will the micro-characteristics of the political system affect policy reinvention by provincial governments in China? What factors influence policy reinvention by provincial governments in China? There is currently a lack of systematic research on these questions.

In China, policy innovation and its “from part-to-whole” diffusion model are regarded as the key way to achieve greater government efficiency [10, 31, 70], and local innovation is an important source of the vitality of reform for the Chinese government [6]. The “One Visit at Most” policy that originated in Zhejiang Province, China, is a typical case of a policy innovation by provincial governments and adopted by the central government. The “One Visit at Most” policy refers to the services directly provided by administrative departments, such as registering real estate, issuing business licenses and personal identification, and having citizens obtain services by interaction with the government once at most or without any physical interaction ([28]:520). According to the official definition, “One Visit at Most” means the government, with the help of the new generation of information technology, optimizes the process of administration affairs among departments and realizes the departments’ power restructuring and process reengineering. The “One Visit at Most” policy is designed to improve administrative efficiency and public satisfaction with the government. Since the Zhejiang provincial government issued the policy in early 2017, Shaanxi, Jiangsu, Anhui, Hubei, and many other provinces have adopted this policy innovation. Investigations into these provinces’ experiences reveals they have commonalities and significant differences in how they conceptualized the problem and their implementation, thus showing the characteristics of policy reinvention during diffusion. It provides rich case material for the study of provincial governments’ policy reinvention in the Chinese political system.

This study focuses on policy reinvention in the diffusion of the “One Visit at Most” policy and explores what factors affect provincial governments’ policy reinvention in the Chinese political system, this expands policy innovation research from focusing on policy invention to including the reinvention of policy followers. This study contributes the unique characteristics of the Chinese system and important theoretical developments to the study of policy reinvention within the broader policy innovation literature.

Literature Review

Over the past half century, there have been more than 1,000 studies on policy innovation and diffusion, most of which have focused on the adoption of policy innovation, the policy diffusion model, and its mechanism. These studies focused less on the content of the policies themselves than the reasons for their diffusion, including time, geographic location, dynamic mechanism, and other factors. Scholars such as Walker (1969) focused on the adoption of policy innovation and have pointed out that whether policy innovations are adopted is determined by intrastate characteristics and information from other states, which provides a heuristic understanding for policymakers [63]. Subsequent studies have noted Walker’s model has dominated research in policy innovation and diffusion for decades [3]. Relatively quickly after Walker’s seminal work, some critics believed the main problem with traditional research on the diffusion of policy innovation is that it ignores changes to the policy innovations that occur after adoption [15, 31]. To some extent, this problem is rooted in dependence on the existing research of event history analysis [60]. In fact, event history analysis models are used to determine whether policy innovation is adopted rather than to examine the diffusion process. Although this method has proved successful in analyzing internal and external policy adoption determinants, it assumes all policy adoption is equivalent in content [56].

Roger first mentioned the concept of policy reinvention in 1978. Since then, many scholars have discussed the issue in the 1980s and 1990s. However, research on policy reinvention has declined since 2000. The related research gradually increased until the past ten years. Policy reinvention is considered a subset of the policy innovation and diffusion life course [8, 45, 47, 62]. In the “innovation” stage, policy followers decide to adopt or reject the new policy; “reinvention” is the process used by policy followers to revise the new policy’s content upon adoption, using knowledge gained from other governments [13, 22, 44]. Policy reinvention transcends the decision to adopt the policy innovation and consists of the process to change the content of the policy itself [29, 30], making it an important issue for research on policy innovation and diffusion.

When discussing policy reinvention, it is necessary to compare it with the concepts of policy innovation and policy invention. Policy innovation means a policy adopted by a local government for the first time, regardless of how long the policy has existed and whether other governments have adopted it [63]. Policy invention refers to the construction of original policy ideas or policy programs, and the application of these ideas or programs is policy innovation [2]. The main object of existing studies is policy innovation instead of policy invention. Policy reinvention refers to the change in innovation modified by adopters in the process of its adoption [48,49,50]. Hays (1996) argues policy reinvention refers to purposeful changes made to innovations as they diffuse [23]. Policy expansion is similar to the concept of policy reinvention. Policy expansion refers to the changes a government makes after it adopts a policy. Policy reinvention is the amendment or change a government makes when it adopts a policy, and there are differences in the timing of the action [4]. Scholars have proposed that policy reinvention can be divided into two types, expanding or reducing the comprehensiveness of policy innovation [20, 44, 56], and considered that the types of reinvention result from characteristics unique to the reinventors [24]. Policy reinvention will vary among different regions and levels of government [5]. As such, this will be a new direction for policy reinvention research.

Scholars have tested numerous factors to determine innovation in a large number of models published since 1990. The determinants can be divided into internal characteristics of local governments and external influences. The two general factors will affect the willingness of governments to innovate [63]. Scholars consider internal characteristics and external influence to be important elements for explaining the behavior of each stage of policy innovation and diffusion, but it is doubtful that all determinants have a stable effect on policy innovation and reinvention [39]. Besides the determinants of policy innovation, some studies posit internal characteristics have a greater influence than external factors on policy reinvention, and internal characteristics dominate the decision of policy revision [8]. Policy followers take into account the policy and the political experience of innovators, allowing them to modify policy contents to align the policy with their preferences [13, 20]. Compared with the research focused on policy innovation, the interstate relationship is a core factor shaping policy reinvention because later adopters can observe the experience of innovators and then change the policy’s content [8]. Additionally, studies have shown policy reinvention is influenced by the characteristics of the national political system, such as the degree of democratization and political elections [55, 59]. According to Shin and Webber (2014), the reasons for changes after adoption mainly consist of ideological differences [53]. The occurrence of policy reinvention depends on the policy position and priority of interest groups, information, and the change in national political characteristics [19, 20]. According to Shin (2010), social and political factors are the main causes of policy reinvention [52]. Policy reinvention may result from the needs and circumstances of the adopters [23]. Lobbying by interest groups can change the motivation of innovation adopters and the content of policy learning [74].

The research on policy innovation and reinvention is focused on democratic and decentralized political regimes. In contrast to federal systems (the separation of powers between the central government and the local government) and democracy (the official election system), China has a specific relationship between the central and local governments and the personnel systems, which has a profound impact on policy innovation and the reinvention of Chinese subnational governments. China’s political system is characterized by a high degree of centralization of political and personnel power and a high degree of decentralization of administrative and economic power [71]. On the one hand, the central government maintains its control over provincial governments through financial centralization and a cadre evaluation system. On the other hand, the central government has given provincial governments a high degree of administrative autonomy to promote economic growth [27, 35], and central–local relations have become characterized by behavioral federalism. Chinese provincial governments enjoy more power than state governments in federal countries [33, 68]. The decentralization of the central government has contributed to China’s rapid economic development and local innovation [17, 66]. In contrast to its counterparts in the West, the nature of policy innovation and diffusion in China is inherently political instead of technocratic.

Recent studies show the driving force of local innovation in China is rooted in the political system [69]. In long-term practice, China has formed a unique model of policy innovation—that is, policy innovation is greatly affected by intergovernmental relations [76, 79]. In China’s unique public policy institutional context, central–local relations are an important perspective for researching policy innovation [42, 80]. Empirical results show the influence that the central government exerts can significantly increase the possibility of provincial governments adopting new policy tools [76]. In terms of horizontal intergovernmental relations, the impact of geographical leadership mobility on policy innovation in China has been a popular topic in recent years. Some scholars regard officials as change agents and point out that when they change their positions, they take policy innovation from their previous stations to the next place they work: portable innovation [73]. Leadership mobility and exchange of officials have a significant impact on policy innovation [38].

To summarize, in the study of policy innovation and diffusion, policy reinvention has become a new research direction. Scholars regard policy reinvention as a phenomenon that occurs in the federal system or decentralized systems, and empirical research has been focused accordingly. In contrast to the federal systems in democracies, China’s political system has specific intergovernmental relations that have a profound impact on the policy reinvention of provincial governments [41, 78]. In recent years, some scholars have paid attention to the diversity of policy innovation results from subnational governments in non-Western countries such as China [12, 37, 75], but research on the factors that influence policy reinvention are sparse. The existing studies are basically exploratory case studies and lack systematic analysis of the factors that influence policy reinvention. The development of theory and empirical research on policy reinvention within the Chinese context requires greater attention.

Analytical Framework

To contribute to the aforementioned research, we analyzed the factors that influence the policy reinvention of provincial governments in China. The framework for this analysis includes two levels: institutional characteristics and internal influences.

Institutional Characteristics

Vertical Intergovernmental Relations

The relationship between a provincial government and the central government is the main factor that shapes policy innovation [25, 26, 75], and it is a key factor in policy reinvention [8]. In China’s semi-decentralized intergovernmental financial relations and official appraisal system, if the central government supports a certain policy, the provincial government is more likely to support the policy as well [25]. Empirical research shows the relationship and interaction between the provincial government and the central government are the driving factors behind policy innovation and learning [75]. In other words, the policy diffusion of provincial governments inevitably will be affected by vertical power relations, particularly policy signals from the central government.

Horizontal Intergovernmental Relations

The policy diffusion process, through which different governments imitate one another, has been an important area of study in public policy research. Relevant research shows provincial governments are more inclined to learn from and imitate their neighboring provinces than distant provinces, and the likelihood of learning and diffusion increases as the distance between the provinces decreases [2, 8, 20]. According to research on the promotion process for provincial officials in China, the comparison of their performance with that of neighboring officials is an important determinant [11]. Therefore, officials may reinvent the policies adopted from neighboring provinces for the purpose of competition [34].

Geographical Leadership Mobility of Governors

In China, local chief officials are the core decision makers of policy innovation. According to Teets and Hurst (2015), senior local officials can influence the policy learning process decisively and interpersonally [57]. Because of the heterogeneity of individual preferences and employment experiences of governors, their geographical mobility will certainly affect policymaking. Researchers have found the activities of policy entrepreneurs contribute to the diffusion of policy innovation [43, 46]. Scholars have studied the role of local leaders’ leadership mobility in promoting the diffusion of administrative reform and believe government officials are policy entrepreneurs who promote policy innovation from within the government [73, 77]. In China, it is normal for the central government to transfer provincial chief officials from one place to another. This horizontal mobility results in governors implementing policies from their former positions into their new posts, which is a policy diffusion process that leads to policy similarity across provinces [78].

Internal Influences

Financial and Technical Capacity of the Government

Scholars have confirmed the impact of local economic development, financial situations, information levels, and other factors on policy innovation [7, 9]. In essence, policy reinvention is a new decision made by the adopter that is influenced by the adopter’s own resource endowment, such as public revenue, organizational resources, and organizational capabilities. Policy innovation usually requires a certain amount of government revenue [14]. Some studies have shown there is a positive relationship between a government’s information technology and its ability to innovate [59]. A government’s technology level affects its reinvention of policy tools.

Heterogeneity of Economic Environment

Existing research on policy diffusion also considers the impact of the heterogeneity of the economic environment among regions [62]. Generally speaking, local governments are more likely to learn from other regions that are similar to their own economic environment, and the characteristics of high homogeneity will make policies easier to learn and adopt. If the heterogeneity among regions is high, then the policy innovation adopted is more likely to be reinvention.

Research Method and Variable Selection

Research Method

Qualitative comparative analysis (QCA) is a case-oriented approach that identifies the necessity and sufficiency conditions to explore “causal recipes” between different combinations of conditions and outcomes, the individual conditions are jointly sufficient to produce an outcome with at least some degree of regularity ([61]:310). The fuzzy-set qualitative comparative analysis (fsQCA) method selected in this study is a combination of qualitative and quantitative methods that can avoid the subjectivity of qualitative research to a certain extent. Scholars now have turned increasingly to fsQCA to conduct small and medium case studies, arguing that it combines the most desired elements of variable-oriented and case-oriented research. The reasons for choosing this method are as follows. (1) Policy reinvention is not the result of a single influencing factor; instead, it is shaped by multiple interacting factors and conditions. QCA can solve research problems centered on (quasi)necessity or (quasi)sufficiency, may stir interest in the causes of a given effect [58], and has an advantage in mining diversified conditional combinations. This is particularly important for explaining the conditions of policy reinvention and the effect of their combinations. (2) The fsQCA method allows for the use of variables that have more than two categories, thus expanding the scope of the phenomenon it can be used to examine.

Variable Design and Indicator Selection

Outcome Variable

Policy reinvention consists of changes that provincial governments make in policy innovations imported from other areas. We used the text content analysis method to analyze whether the adopters reinvented policy. The entire process of government activities generates policy texts; they are a natural imprint left by governments in the process of dealing with public affairs. Therefore, there is a certain rationale to choosing policy texts to measure policy activities, including policy formulation and implementation. Further, it is convenient to make an objective comparison of all provinces at the same level. However, we should note the measurement of outcome variable has defects, and policy texts cannot be representative of all reinvention practices.

We collected the “One Visit at Most” representative policies of each province. The policy selection criteria are as follows. (1) Policies are issued by provincial (province-level municipality) government departments from 2017 to 2018. (2) It is clearly stated in the policy documents that the objectives and measures are working to implement the “One Visit at Most” policy. After screening the policies, each province’s policy contents are analyzed to identify reinvention measures. To further test whether these measures are genuine reinventions, we have analyzed the official reports from each province to verify whether these policy measures have been implemented. The analysis results show that in the case of the “One Visit at Most” policy, twenty-seven provincial governments have adopted the policy, and fifteen of the twenty-seven provincial governmentsFootnote 1 have carried out policy reinvention in practice. The fifteen provincial governments have their own unique reinvention measures compared with Zhejiang Province and provinces that adopted the innovation earlier. Provinces with reinvented policies were assigned a value of 1; otherwise, they were assigned a value of 0.

Conditional Variable

According to the QCA standard practice for medium-sized samples (ten to forty cases), the number of conditional variables should be between four and seven [1]. The number of conditional variables used in this paper falls within this range. According to the aforementioned analysis framework combined with the “One Visit at Most” policy reinvention practice, this paper chooses six conditional variables. All of them, or their combinations, may play an important role in the explanation of policy reinvention (Table 1). This study uses the fsQCA calibrate function to calibrate continuous variables based on three-value fuzzy sets [1]. This function requires three values of the variable as anchor points: 1 (fully in the set), 0.5 (the crossover point), and 0 (fully out of the set). For the category variable, this study uses a two-value set with a membership score of either 0 or 1 (either in or out of a set; see Table 2).

Table 1 Variables and indicators of measurement
Full size table
Table 2 Set membership of respective conditions and outcome after calibration
Full size table

Policy Signals Released by Central Government (PSRCG)

Based on the perspective of vertical intergovernmental relations, we selected the policy signals released by the central government as a conditional variable. In March 2018, the Central Leading Group for Comprehensively Deepening Reforms strongly recommended promoting the “One Visit at Most” policy to the whole country. After that, the policy was widely adopted at the provincial government level. In this study, provinces that adopted the policy before March 2018 were assigned a value of 0, and provinces that did not adopt the policy were assigned a value of 1.

Geographical Leadership Mobility of Governors (GLMG)

From the Database of Local Leaders, China, we collected the work experience of leaders in the twenty-seven provinces that adopted the “One Visit at Most” policy. Whether the provincial and deputy provincial governors had work experience in Zhejiang Province in the five years before 2017 was classified as a category variable. When provincial and deputy provincial governors had served in Zhejiang Province in the five years before 2017, the variable was coded 1; otherwise, the variable was coded 0.

Proximity to the Innovating Province (PIP)

From the perspective of horizontal intergovernmental relations, the geographical distance between each provincial capital and the Zhejiang provincial capital is used as the category variable to measure the difficulty of diffusion and the driving force of policy reinvention. The closer a province is to Zhejiang, the more likely it is to be motivated to reinvent policy. We calculated the geographical distance between Zhejiang Province and other provinces (the capitals of each province). The results were calibrated using fsQCA software according to the three-value fuzzy set.

Heterogeneity of Economic Environment with the Innovating Province (HEEIP)

This paper measures the heterogeneity of the economic environment between each province and Zhejiang Province by using the Marketization Index (MI) of provinces in China for 2017. This index is selected because it reflects the level of regional economic development among provinces. In this paper, the indicator is used to measure the absolute value of the difference in the MI between each province and Zhejiang. The higher the HEEIP, the more likely it is to reinvent policy. The results are calibrated in fsQCA software according to the three-value fuzzy set.

Regional Internal Financial Resources (RIFR)

For provincial governments to implement the “One Visit at Most” policy, they must have sufficient financial capacity. In this study, per capita fiscal revenue for each province is taken as a continuous variable. In the fsQCA, the results are calibrated in fsQCA software according to the three-value fuzzy set. Provinces with sufficient financial capacity are more likely to reinvent policy.

Informatization Foundation (IF)

The “One Visit at Most” policy requires that the provincial government has sufficient information technology ability. Whether implementation involves the construction of an e-government platform or the disclosure and sharing of information, certain hardware and software skills are needed. In this study, the Information Development Index (IDI) of each province is selected to measure the level of information technology ability. The IDI is used to measure a region’s informatization infrastructure construction, informatization application level, and restriction environment. The data are taken from the Evaluation Report of China’s Information Development, issued by the China Electronic Information Industry Development. Provinces with a sufficient information foundation are more likely to reinvent policy. In fsQCA, the results are calibrated in fsQCA software according to the three-value fuzzy set.

Empirical Analysis Based on fsQCA

Necessity Conditions Analysis

In this study, we used fsQCA3.0 software to analyze the necessary conditions for policy reinvention after calibrating the output truth table of fuzzy sets for the relevant continuous variables. The purpose of necessity condition analysis is to identify the conditional variables that have a major impact on the outcome variable and to conduct in-depth research on the remaining conditional variables through conditional combinations. Table 3 shows the results of the necessity analysis of the conditional variables.

Table 3 Necessity test results for conditional variables (Value of Result Variables = 1)
Full size table

In a necessity analysis, consistency represents the degree of influence of conditional variables on whether the outcome variable has a value of 1—that is, to what extent a value of 1 on the outcome variable requires a certain value on the conditional variable. Generally, the criterion for necessary conditions consists of having a consistency score of more than 0.9. In Table 3, the consistency of the variable PSRCG is 0.933333, which meets the criterion of the necessary conditions. The other variables do not meet the criterion of the necessary conditions, but their combination effects can be analyzed.

Conditional Combination Analysis

Conditional combination analysis measures the effects of different combinations of conditional variables on the results when a single conditional variable does not constitute a necessary condition. In conditional combination analysis, to avoid the influence of necessary conditions on the outcome variables, the necessary conditions are eliminated, and the remaining five variables are analyzed by conditional combination analysis. In the fsQCA software, the outcome value is set to 1, and the key consistency values are set to 0.8 and 0.9. In Table 4, the results are shown using the following symbols: ● indicates the occurrence of the precursor condition; ⊗ indicates the precursor condition does not appear; and a blank space indicates the precursor condition may or may not appear.

Table 4 Conditional combination analysis results (Outcome Variable = 1)
Full size table

There are two paths for a conditional combination when the consistency key value is set to 0.8: (1) GLMG * PIP * ~ HEEIP * ~ RIFR* ~ IF and (2) GLMG * PIP* ~ HEEIP * RIFR* IF. After increasing the consistency key value to 0.9, the second combination path under the consistency cutoff of 0.8 is the only path remaining. The combination of these three conditions in the expression of Boolean algebra can be simplified logically as follows: GLMG * PIP* ~ HEEIP. These three conditional variables contribute most to policy reinvention. To test the robustness of the results of conditional combination analysis, we carry out conditional combination analysis for the case with the outcome variable set to 0—that is, the analysis of the cases with no policy reinvention after the adoption of policy innovation. The results are shown in Table 5.

Table 5 Robustness detection for conditional combination analysis (Result Variable = 0)
Full size table

The analysis results in Table 5 show the combination of conditions for the absence of policy reinvention outcomes remains unchanged after increasing the key consistency values from 0.8 to 0.9: ~ GLMG * ~ PIP * HEEIP * RIFR. The results of Tables 4 and 5 show the geographical leadership mobility of governors, the proximity to the innovating province, and the heterogeneity of economic environment with the innovating province are the main conditional variables affecting policy reinvention.

Conclusion and Discussion

Conclusion

Through the fsQCA analysis, the following conclusions can be drawn from the analysis of necessary conditions and condition combinations.

First, policy signals released by the central government are a necessary condition for the provincial government’s policy reinvention. China’s political system does not have a decentralized federal structure. However, it does have some characteristics of decentralization under a strong central authority. The central government has a considerable role in promoting the diffusion of provincial governments’ policy innovations. When the central government releases a strong policy signal, provincial governments have a strong incentive to carry out the policy innovation and will respond to the signal. The results of this study indicate the administrative pressure of superiors is a necessary condition for the “One Visit at Most” policy reinvention in China.

Second, geographical leadership mobility of governors, proximity to the innovating province, and heterogeneity of economic environment with the innovating province are the main factors affecting policy reinvention. The results of this study show that having provincial leaders with prior experience in the Zhejiang Province, proximity to the Zhejiang Province, and economic environments similar to the Zhejiang Province are the core sufficient conditions for provincial governments to carry out policy reinvention. Previous studies generally have shown the mobility of local leaders leads to policy isomorphism. Notably, a similar economic environment may lead to policy learning and adoption, but it may also lead to policy reinvention.

Overall, the results verify the characteristics of political systems in China are important factors in the subnational government’s policy reinvention. Intergovernmental relations and the cadre system with Chinese characteristics profoundly shape the reinvention ability of governments. Policy reinvention by provincial governments is not the result of a single factor. In China, vertical intergovernmental relations (that is, the policy signals sent by the central government in the political system) are the necessary condition to influence the provincial government’s policy reinvention. The combination of factors for horizontal intergovernmental relations (the proximity to the innovating province), the geographical leadership mobility of governors, and the characteristics of the government’s own economic environment will affect the provincial government’s policy reinvention. In contrast to previous studies, provincial governments’ internal characteristics (their financial and technological capabilities) may not be sufficient conditions for policy reinvention in China. This may explain that the determinants of reinvention may play diverse roles in different institutional backgrounds.

Discussion

Understanding policy reinvention is of great significance to the study of policy innovation and diffusion. To some extent, policy diffusion allows a degree of flexibility among subsequent adopters. Policy reinvention shows how governments adapt and change when the diffusion of policy innovation occurs. Time-ordered diffusion means only early policy adoption is innovative, but reinvention means the governments that adopt policies later may also be innovators. Policy reinvention provides another perspective from which to consider the policy innovation of subnational governments. Although provincial governments will adopt the basic idea of a certain policy, this research on policy reinvention shows provincial governments that adopt policy innovations will not blindly follow the prototype and may be influenced by institutional factors.

This study expands the traditional policy innovation and diffusion literature by adding policy reinvention as a dependent variable. We construct an interpretative framework for the results of policy reinvention, which is in line with the new direction of current policy innovation and diffusion research. In China, innovation is regarded as a key way to achieve higher governmental efficiency from central and local governments [10, 32, 36, 70]. Conventional studies on policy reinvention have focused mostly on Western democratic countries, yet we can verify the existence of policy reinvention in China and contribute fresh empirical findings on policy reinvention by examining the institutional and internal factors that influence the provincial governments’ policy reinvention. This study serves as a window for researchers to explore larger issues about policy reinvention. Additionally, this study uses fsQCA to conduct cross-case studies and promotes the validity and reliability of research conclusions based on previous single-case studies.

The shortcomings of this study are as follows. First, the fsQCA method selected has inherent defects, such as the use of calibration when dealing with continuous variables that may involve subjective factors. Additionally, QCA cannot fully reflect the causal relationship and does not fully explain the underlying causal mechanism [51]. Relevant conclusions can be supplemented by other methods, such as comparative case study, to provide more powerful explanations for relevant issues. Second, this paper discusses the reinvention of only one policy among different adopters, so the applicability of the conclusion has limitations. Determining whether the analytical framework and conclusions are applicable to other policy areas requires further study.

As an exploratory study, what we must emphasize is that our research is not committed to providing a comprehensive explanation for policy reinvention. The appropriateness and explanatory power of this framework must be further verified and revised based on the development of future research hypotheses. Future research can continue to explore the micro basis of policy diffusion and reinvention in different institutional contexts. Comparative studies or case process tracking can be used to supplement the causal mechanism analysis to explore the causal relationship, and this can be verified in other policy areas. In addition, researchers can consider the policy attributes and characteristics in the policy reinvention analysis framework or take the policy reinvention types as the research object, then analyze the influencing factors of different types of policy reinvention.