Introduction

In the past decade, the literature on policy piloting has grown exponentially in the China study and political sciences’ fields; most of these articles acknowledge the success and benefits of experimental governance (Checkland et al. 2023; Ettelt et al. 2022; Huitema et al. 2018; Heilmann 2008; Zhu and Bai 2020). China has garnered increasing attention due to its success in experimental governance, particularly since the economic transformation that began in the 1980s (Rawski 1995; Xu 2011a). The Chinese governance model, with policy piloting as one of its core principles, has demonstrated remarkable adaptability and flexibility in promoting economic growth and technological development. Recently, Heilmann (2018) referred to it as a “black swan” for its unexpected yet significant impact.

However, few studies have paid attention to the drawbacks and failures of experimental governance (Zhu and Zhang 2020), particularly the processes and political-economic dynamics from the perspective of local governments, which ultimately decide whether and how to enact policy innovations at the localities. Scholars argue that local governments are typically motivated to participate in central policy pilots that involve fiscal transfers or resource allocation from the central government. However, limited attention has been given to policy pilots that lack central funding and resources (Gong 2022; Heffer and Schubert 2023; Wang and Yang 2021). In such cases, the capacities of local governments become critical (Jia et al. 2014; Teets et al. 2017), significantly influencing their strategies for responding to central policy piloting objectives. This research aims to address this gap by investigating how local government capacity shapes their strategies in responding to central policy pilots without fiscal transfers or resource allocation.

This article seeks to identify the factors that determine local governments’ innovation strategies when responding to central policy piloting objectives without central funding or resource allocation in the experimental governance process. This research builds on recent developments in the scholarship on local governments’ responses to central piloting goals (Guo et al. 2021; Ma and Liu 2024). In comparison, this study makes several new contributions to the field. First, it broadens the literature by revealing diverse response strategies adopted by local governments, including creative, symbolic, and distorted policy innovations, thus providing insights into central-local dynamics in experimental governance from the perspective of local governments. Second, this research emphasizes the political-economic dynamics of local governments’ responses to central piloting goals, particularly highlighting the role of local government capacity, which has been largely underestimated in the existing literature.

The main goal of this article is to unfold the black box buried within China’s large-scale policy piloting process, from the perspective of local governments, while illustrating why and how distinctive regional economic conditions and central-local dynamics matter in shaping experimental governance at the local level. Empirically, this article posits particular attention to China’s National New-Type Urbanization Plan launched in 2014, which jump-started several rounds of local policy testing, through which the Chinese government aims to push forward the reform of the hukou system (urban household registration system), as well as accomplish a targeted goal of turning an additional 100 million urban residents into urbanites between 2014 and 2020.Footnote 1 To describe and theorize the different policy responses at the local level, we adopted a comparative-case analysis approach and interviewed 55 local governmental departments, engaged in policy piloting across nine city-county level governments in provinces H and J.

This article is organized as follows. The second section reviews the role of local governments in experimental governance and the political-economic dynamics of their responses to central piloting goals in China and beyond. It also constructs a theoretical framework to guide the case study. The third section provides an overview of the implementation of China’s New-Type Urbanization Plan through policy piloting. The fourth section details the research methodology and data collection methods. The fifth section presents the case study, examining how local government capacity shaped differentiated responses to central piloting goals. The sixth section discusses the broader implications of the findings, particularly in relation to the costs of experimental governance. The final section concludes the study, offering policy implications and outlining research limitations.

Exploring local governments’ response strategies to central piloting goals

The role of local governments in the policy piloting process

Policy piloting under hierarchy involves both central and local actors. Two basic approaches, towards the political dynamics of policy piloting in China’s governance structure have emerged, fundamentally debating the role of central and local governments in the country.

One approach emphasizes the comparative advantages of both the central and local governments, highlighting local governments’ role in finding suitable policy instruments that the central government needs. This approach is best embodied in the “experimentation under hierarchy” model developed by Heilmann (2008), in which the central government allows space for locally initiated policies and encourages local governments to explore policy options through experimentation. Under a decentralized experimental governance structure, local governments are incentivized to participate in national policy piloting as a mean to promote local change and address local problems (Ettelt et al. 2022). By conducting policy “experiments in accordance with local conditions”, they can enhance local governance, while simultaneously providing the central government with potential policy instruments for future nationwide implementation. Zhu and Zhao (2021) further developed the dominant model by studying local governments’ authorized autonomy in forming piloting goals and instruments.

The conscious policy design approach, however, embraces the decisive role of the central government in controlling the piloting process and variables intentionally, while acknowledging and enabling a relatively marginal degree of local autonomy in experimental governance, for local authorities. In other words, experimental governance is taken as a policy tool to illustrate the ruling party’s attitudes and preferences towards policy change or typically as an instrument of persuasion (Ettelt et al. 2015a). Evidence for this argument was provided concerning the urban housing reform (Mei and Liu 2014); the local political reform (Tsai and Dean 2014); the piloting process regarding “eco-city” projects (Miao and Lang 2015); the New Rural Cooperative Medical Scheme (Song and Li 2024); as well as in the agricultural tax reform (Wang 2019).

Both approaches highlight the necessity to concentrate on the diversified roles of the central government, while underestimating how local governments’ behaviors cannot be understood as a unitary model as these vary across localities while piloting with policies. The dominant explanations assume that local governments will respond to the center positively and always offer the best policy instruments, in accordance with their respective goals. In reality, local governments have to take the interests of the center, of local citizens as well as their own self-interests into consideration before deciding how to respond to these strategies (Heffer and Schubert 2023; Tsai and Dean 2014; Zhu and Zhao 2021). Evidence from the UK also suggests that policy pilots are not always effective (Ettelt et al. 2015b).

Local governments’ behaviors have been frequently studied in policy innovation literature. Teets and William (2015) argue that local governments enjoy relatively high levels of autonomy in local policy innovation. Local officials are seen as policy entrepreneurs who have incentives to innovate policies due to the enticements derived from cadre promotion and local needs. Teets and William (2015) group local governments’ policy innovation strategies into four categories, including compliance, resistance, innovation and involution, and then argue that policy innovations are not always conducive to better local governance. Specifically, in policy piloting, the difference is that local governments have no specific policy options to comply or resist but have to positively or negatively innovate policy instruments to respond to the central piloting goals.

Local government capacity and response strategies to central piloting goals

When examining local governments’ response strategies to central piloting goals, it is essential to adopt the “incentive-capacity” model to explain their innovative behaviors within experimental governance.

  1. (1)

    Incentive Factor. Within China’s central-local bureaucratic system, the central government initiates policy piloting programs, while local governments compete for the opportunity to serve as pilot sites. This creates an interaction between the central “top-level design” and local “autonomous exploration”. Local officials often respond proactively to the central government’s calls for policy piloting. On the one hand, these pilots typically receive financial support from the central government or higher authorities to ensure smooth implementation and promotion at the local level (Heffer and Schubert 2023; Zhu and Zhao 2021). On the other hand, policy pilots provide local officials with a platform to showcase their achievements and demonstrate loyalty to the central government, as successful pilots can aid in their career advancement (Xu 2011b). Scholars argue that local innovators remain motivated by institutional incentives, political pressures, peer influence, and policymakers’ risk-taking personalities (Hasmath et al. 2019; Teets and Reza, 2020).

    However, local governments’ responses to central policy experiments vary (Ma and Liu 2024). Local governments possess multiple roles: as agents of the central government, they are required to comply with higher-level directives in managing local affairs. Simultaneously, as rational economic actors, they seek to maximize local political and economic interests. Influenced by both the promotion mechanism and the financial contracting system, policies that contribute to fiscal growth are more likely to gain local government support (Ye and Wu 2024). Additionally, as social actors, local governments are tasked with serving the local public, making policies that attract significant public attention more likely to be actively implemented (Guo et al. 2021).

  2. (2)

    Capacity Factor. Local governments operate with significant autonomy under central government supervision (Oi 1992; Jin et al. 2005). When determining their level of enthusiasm for applying to policy pilots and the extent to which they fulfill these tasks, local governments weigh several factors. These include the alignment of their interests with central policy goals, the authority of the policy initiator, and the specific objectives and fields of the policy pilots. Additionally, local governments must assess their available resources and the costs they can bear, which depend on the region’s GDP, fiscal revenue, infrastructure, and other objective conditions (Tsai and Dean 2014), as well as the feasibility of advancing the policy pilots. Limited resources at the local level and flexibility in policy implementation are key to understanding the differentiated responses of local governments to central policy pilot tasks.

Although pilots are often seen as capacity-building processes (Qian 2017), existing theories have not sufficiently addressed local governments’ innovation capabilities or the economic and institutional constraints they face. Some scholars have highlighted the importance of local governments’ fiscal and administrative capacity in supporting innovation (Qian 2017; Zhu and Zhao 2021), aligning with the organizational level of policy capacity (Wu et al. 2015). However, these studies have failed to connect these capacities to the mechanisms that produce symbolic or distorted strategies, which in turn weaken local governance.

Overall, after assigning piloting goals to local governments, the central government grants them considerable autonomy to engage with new policy instruments within their specific contexts. Local government capacity and implementing pressures play a crucial role in shaping their behaviors and response strategies to central piloting goals, especially in the absence of central fiscal transfers and resource allocation. We have developed a theoretical framework to better understand the logic behind local governments’ differentiated response strategies to central piloting goals within a hierarchical system, as illustrated in Fig. 1.

  1. (1)

    Local Government Capacity. This refers to the ability of local authorities to effectively perform their functions, manage resources, and deliver services to their communities. Simply put, a local government’s capacity is its ability to do what it wants to do (Gargan, 1981). Among the various dimensions of local government capacity, financial resources and innovative capabilities are particularly important. In China’s multilevel bureaucratic system, administrative rank is a key indicator of local government capacity, as fiscal and policy resources are allocated vertically. Particularly, in the absence of central policy support during the piloting process, local government capacity becomes the sole organizational resource.

  2. (2)

    Implementing Pressure. Implementing pressure refers to the dynamics and influences that compel local authorities to effectively execute pilot policies and initiatives. Local governments’ implementing pressure is primarily related to whether the central piloting goals are overly ambitious and whether they have prior experience in the specific policy field. The higher the central pilot target and the weaker the local government’s piloting experiences, the greater the implementing pressure it faces.

    Altogether, these two factors shape local governments’ response strategies to central pilots, which can be either creative or, conversely, symbolic and distorted. Creative Innovation refers to the development of genuinely new ideas, products, processes, or policies that offer novel solutions or improvements. Symbolic innovation involves changes that are more superficial or representational rather than functional. Distorted innovation refers to changes that are intended to be creative but end up being misaligned with their intended goals. When local government capacity is strong, local authorities tend to adopt innovative strategies to meet the policy objectives of the central government and address the needs of local residents, even when facing high levels of implementing pressure. Conversely, if local government capacity is weak, local authorities are more likely to adopt symbolic strategies under low implementing pressure, and distorted strategies under high implementing pressure.

  3. (3)

    Local Governance. Local governance is defined as collaborative, hybrid forms of interaction between government, market actors, and civil society. It is seen as a response to the growing complexity of societal problems, where government alone is not the sole source of expertise and authority (Røiseland 2011). We argue that not all the instruments adopted are conducive to the improved local governance, especially when a policy instrument is deviated from local needs for the following two reasons: first, local policymakers attempt to maximize their self-interests; second, policymakers seek to fulfill the piloting goals from the centre rather than local residents’ demand because these are appointed by upper-level officials and not by citizens. As a result, the adoption of some of the instruments is symbolic or distorted and could be harmful to local governance, given that local needs are scapegoated for policymakers’ self-interests.

Fig. 1
Fig. 1
Full size image

Framework of local governments’ differentiated response strategies to central piloting goals under a hierarchy system.

China’s urbanization reform through policy piloting

The last four decades have witnessed China’s rapid economic development and urbanization process, during which social actors’ interests increasingly diverged, thereby creating enormous challenges for the government and the implementation of social reforms. Among them, the hukou system (the Chinese household registration system) is the most controversial issue (Heilmann 2016). The central government prefers a gradual approach through which it can control the pace of the reforms, in order to increase population mobility across regions while guaranteeing social stability. More importantly, the conflicts regarding the reforms in China’s hukou system are generated by the differing interests of urbanites, rural residents, migrants, and local governments, especially in megacities. Under such scenarios, local governments have increasing difficulties in managing relations with different social actors and in pushing forth relevant reforms.

The Chinese government adopted a so-called “experimental governance approach” for selecting better policy instruments to push forward these reforms. In 2014, the State Council formulated the “New-Type Urbanization Plan - NUP - (2014–2020)”, aiming to urbanize hundreds of millions of rural residents and migrants by 2020. The plan created massive pressure on local governments to expand public goods provision, and to accelerate the reform of the hukou system and other related economic and social systems. This urbanization plan was first implemented on an experimental basis, involving the majority of local governments in China.

During this process, the National Development and Reform Commission (NDRC), in its capacity as the central policy coordination department, formulated several pilot programs and determined three batches of comprehensive pilot regions in 2014, 2015 and 2016, with a total of two provinces and another 232 cities, counties and towns. At the national level, the urbanization plan includes a series of specific pilot tasks, and each pilot region, including city and county governments, takes responsibility for different tasks. Among all the pilot tasks, the reform of the hukou system was of crucial importance and assigned to all the piloting localities. After two years of policy piloting at the local level, in 2016 the NDRC sent out investigative teams to evaluate the local performance in terms of policy innovation. The teams then selectively fed some policy instruments back to the national level at the end of 2016, and then eventually the policy instrument was implemented nationwide.Footnote 2

It is interesting to examine how local governments respond to the policy piloting goals of the central government. In the case of the NUP, it turned out to be a project that grants high level of local autonomy without providing a substantial financial support from the centre. According to the internal policy document the NDRC issued to the investigative team, “local governments can seek for central supports regarding any policy but not for the fiscal and urban construction land policy”. Especially for those localities lacking sufficient fiscal revenues, it thus becomes very difficult to serve as testing grounds. In the words of a provincial coordinating officer:

When it was time to select the pilots, our local government cadres were enthusiastic about applying; but after the central government approved the bid, it became very hard for the local governments to accomplish the tasks within a short time span.Footnote 3

By taking into considerations the huge variations across the sub-national level in China, this article will next investigate how local governments respond to the central experimental goals when they do not receive any support from the center; what are the different strategies adopted in policy instruments innovation; whether these differences are directly dependent on the local political-economic conditions.

Research methodology and data collection

In this article, we adopted a qualitative research method, based on a comparative case study approach at the county-or-city level; we selected nearly all the city and county governments, which were assigned different pilot tasks in provinces H and J in China. These two provinces are representative for mainly two reasons: one is that they were all assigned pilot experiments for the hukou system, which helps us compare the responding strategic differences at the subnational level when referring to the same reform; the other reason is that the two provinces differ in terms of geographical location as well as in levels of economic and social development, thus proving our point that these characteristics do play an influential role in policy pilots and their success (or lack thereof). Specifically, the economy in Province J, located in the southeast of China, is highly developed, while the economy in Province H, located in the north of China, is less developed.

As for the data collection, two of the authors were authorized to join the national investigation team, which provided a unique and exclusive opportunity to conduct intensive fieldwork in both provinces from August 15 to August 20 in 2016 and from August 28 to September 03 in 2016, respectively. Through the interviews with a dozen of local political leaders we were able to obtain several internal documents and information, providing a detailed indication of their “insider’s” perspective. Specifically, we investigated all the pilot regions in Province H, including cities S, D, W, Z and B. In Province J, given that the entire province is considered a pilot region,Footnote 4 we selected four cities with different economic conditions and geographical characteristics but similar pilot tasks, including cities NJ as the provincial capital, WX, NT, and XZ, as well as some counties or towns within their jurisdictions. Several basic socioeconomic indicators of all the survey regions are shown in Table 1. We define strong local government capacity as a region where economic development and fiscal resources exceed the national average, whereas weak local government capacity refers to regions with below-average levels of economic development and fiscal resources.

Table 1 Comparison of local government capacities in provinces H and J, 2015.
Full size table

We adopt a meso-level analysis on the behaviors of local governments’ departments (Shin 2017). In each region, we conducted a symposium and interviewed the leaders of relevant government departments, responsible for policy piloting under the national urbanization plan. Some of the departments interviewed are listed in the appendix section. We carried out semi-structured interviews with 55 local leaders and departments’ officers, with each interview lasting approximately 2.5 h. For ethical consideration, our respondents’ names have all been anonymized (Mele et al. 2020). In addition, based on a first-hand investigation, we also collected secondary literature, including policy documents and survey reports provided by local governments as well as media news reports.

What shapes local governments’ response strategies to central policy goals in China

Central piloting goals and implementing pressure on local governments

The central government’s goals regarding the hukou system reform hinge on two tenets. The first is to push forward the urban residence permit reform to equalize the availability and offer of public services’ provisions for both migrants with an urban residence permit as well as for those with a permanent urban hukou so that the former group would de facto be entitled to full “citizenship” rights. Since 2010, a resident permit policy has been implemented nationwide, which served as a temporary substitution to delay the hukou reform: on the one hand, the resident permit provides basic urban public services for migrants, but at the same time the resident permit policy does not place too much financial burdens on city governments in comparison to the hukou system or to full equalized rights for rural migrants.

Under this framework, the central government’s desired goal to reform the residence permit is meant to allow urban migrants or settlers to enjoy (nearly) the same public services as those enjoyed by urban hukou holders, including the same education, medical treatment, social security, and pension. However, the central policy piloting goals for urban residence permit reform were vague, lacking clear quantifiable indicators. For example, the piloting goals in City S stated that “Promote equalization of basic public services and significantly improve the level of public services.”Footnote 5 This suggests that the piloting goals for urban residence permit reform did not place significant pressure on local governments during China’s NUP.

Meanwhile, the second piloting goal is to accomplish the envisioned urbanization rate by including those rural-to-urban migrants with a resident permit as well as urban citizens with urban hukou permits. Starting in 2012, under the new administration of Xi Jinping, the urbanization process has been considered an important means to boost domestic consumption while at the same time helping the Chinese economy transform from an investment-driven to an innovation-driven system (Andreas and Zhan 2016; Ye and Wu 2014). Therefore, urbanization rates are taken as the basic indicators to measure the level of urbanization so far achieved in China. For example, the central government set a hukou permits urbanization rate of 62 percent and a resident permits urbanization rate of 68 percent for Province J by the year of 2017, respectively; while the rates in 2013 were 57 percent and 64 percent, respectively. Similarly, the piloting goals for City S, the capital of Province H, are envisioned urbanization rates of 48 and 60 percent by the year of 2017, respectively; while the rates in 2013 were 41.47 percent and 54.96 percent, respectively.Footnote 6 Table 2 lists the piloting goals for urbanization rates in 2017 (compared to the rates in 2013) for all piloting regions in provinces J and H.

Table 2 Urbanization rates in 2013 and expected urbanization rates in 2017.
Full size table

According to the theoretical framework, we argue that implementing pressure shapes local governments’ response strategies in fulfilling central piloting goals. Generally, from the perspective of local governments as policy pilot implementers, there are two major challengers in meeting the central goals. First, broadening the range of public services attached to the resident permits would result into a massive increase of public fiscal expenditures for local governments, most of which are already faced with strict budgetary constraints, especially regarding the welfare services’ provision of education, health care, etc.Footnote 7 Second, to a certain extent, the central government’s two basic goals turn out to be contradictory: improving the benefits of resident permits could increase the migrants’ urbanization rate, while hindering the growth of urbanites’ hukou urbanization rate, given that the differences between the resident permit and the urban hukou are getting smaller. Altogether, both problems increase the difficulties for local governments to innovate new policy instruments.

In comparison, the piloting goals for urbanization rates placed significantly more pressure on local governments than the urban residence permit reform, as local implementers were required to meet quantifiable targets by the end of 2017. In the following sections, we will demonstrate how varying local government capacities and levels of implementing pressure shaped the diverse strategies employed by local actors in response to central piloting goals, both with and without quantifiable indicators.

Local government capacity and response strategies to central piloting goals

Weak local government capacity, differentiated implementing pressure and response strategies in Province H

Overall, local governments in Province H had weaker government capacity and they adopted the reform symbolically and have made little progress regarding the resident permit reform. Although the public service items attached to the residence permit have increased, the actual improvements on individuals’ rights remain limited. As a basic framework, the Provincial H Government promulgated the “Residence Permit reform Measures of Province H (Trial)” on February 15, 2016, in which the new residence permit contains 5 kinds of rights and 8 basic public services.Footnote 8 Based on this framework, local governments can improve on the contents of the resident permit; otherwise, they just conform to the basic level. City S, the capital, is the only place that made some improvements over the provincial standards. Specifically, the government entitled public education rights to the migrants and local peasants’ children; under these new scheme 87.4 percent of migrants’ children have enjoyed rights to public education. Comparatively, local governments in Province J have entitled urban migrants with public education rights since the year of 2005 and nearly 99 percent of migrants’ children have enjoyed these rights for quite a long period.Footnote 9

Interestingly, the resident permit reform has a negative effect on the hukou urbanization rate policy piloting, reflecting a high level of implementing pressure in fulfilling urbanization rates policy goals. For local governments in Province H, pushing rural residents to settle in a city and apply for an urban hukou is not an easy job, despite governments having eliminated all the entry barriers for migrants and local peasants. The city S, capital of Province H, for instance, has a “zero threshold”Footnote 10 to get an urban hukou permit.Footnote 11 In theory, the peasants’ willingness to settle down in the city and apply for an urban hukou mainly depends on the balance between the attractiveness of an urban versus a rural household registration. However, in recent years, the attractiveness of retaining a rural household registration has increased rapidly, due to an increase in both public services provided to those with a rural household registration and the potential value of rural lands (including rural homesteads and village collective land). As a result, the peasants’ willingness to settle in cities and apply for an urban hukou has been greatly reduced, which makes it difficult for local governments to meet the center’s goals regarding its new urbanization policies. As one local cadre in Province H expressed, the implementing pressure was exceptionally high:

The pressure derived from the goals of quickly urbanizing the country is heavy. The Provincial H Government was assigned an urbanization rate goal of 60%, while City B was tasked with a rate of 50%. These national assessment indexes inflict great pressures on local governments at all levels.Footnote 12

Although the Provincial H Government reformed the household registration management system to provide easier access for migrants and peasants applying for urban hukous, the success rate is quite low. A local cadre interviewed in Province H cited an internal survey on peasants’ attitudes towards migration, stating that, in the aftermath of the new migration registration system being established,Footnote 13 the peasants’ willingness or unwillingness to settle in a city were 31 and 33%, respectively. The City S Government faced the same situation; there were only 30 thousand additional urban hukou applicants per year, since the new policy was established in 2015.Footnote 14 The only case that presents a different story in Province H is City B, a famous commercial center in the North of China; here the number of migrants reached 170,000 but the population with an urban hukou only amounted to 62,800 in 2016. When the City B Government decreased the requirements for applying hukou permits, nearly five thousand migrants came into possession of a regular urban hukou permit in the first half of 2016 only.Footnote 15

Apart from this successful case, we find that local governments in Province H tend to adopt partial policy instruments to meet the central piloting goals as quickly as possible, rather than creatively promote local governance through policy piloting. We found that local governments with weaker capacity tend to adopt distorted policy instruments. For example, local governments in Province H are fond of pushing forward the “town-district conversions” and “county-district conversions” reform in the name of “expanding urban development space”, which can improve the urbanization rate quickly by a simple change in nomenclature. In fact, through the above administrative reorganization reforms, once the land is “re-converted” the peasants would automatically change their rural residents’ status, thus increasing the urbanization rate. According to our survey, three counties in City S’s jurisdiction had adopted the “county-district conversions” reform in 2014, which rapidly increased the urbanization rate (on paper). As one of the cadres in Province H said:

According to our “City 13th Five-Year Plan”, we need to accelerate the “town-district conversions” reform “by changing the regions located in the county government and surrounding towns to the districts of city”. This work is the most vigorous and effective for policy piloting.Footnote 16

Although distorted innovation may help fulfill central quantifiable goals under high implementing pressure, it can undermine local governance at the expense of citizens’ interests. This way of carrying out the policy goals in Province H increases the fictitious rate of urbanization quickly, thereby allowing localities to meet the goals of the central government. However, these result into the peasants bearing the costs of such slanted policies. In reality, the fact that peasants become urban citizens virtually overnight is merely due to the changes in the statistical caliber,Footnote 17 which ultimately does not allow these citizens to get an urban hukou, nor enjoy the same urban public services. Furthermore, this “leeway” while allowing local governments to meet central goals, does not empower local citizens by giving them better and improved services or higher purchasing ability, which is ultimately one of the desired goals of the center. Even more dramatic is that afterward these individuals shall not be classified in governmental statistics as “peasants” any longer, thereby losing their rural residents permit, as well as their rural collective assets. Between 2014 to 2016, several administrative reorganization reforms frequently took place in Province H.

Strong local government capacity, differentiated implementing pressure and response strategies in Province J

In comparison, local governments in Province J have higher capacities than those in Province H, as shown in Table 1. However, local governments in Province J also faced high levels of implementing pressure, particularly regarding the urbanization rate piloting goals. Specifically, the urbanization rate in Province J was very slow and much lower than the original estimated goal set by the central government in recent years. According to an official survey carried out in Province J, peasants’ willingness to settle in cities was at 20.1%, which is even lower than the rate in Province H. In fact, a counter-urbanization phenomenon has recently emerged in Province J, with some urbanites wanting to change their residence permit back to a rural one.Footnote 18 More and more migrants and peasants move back to live in rural areas, due to the increasing attractiveness of rural households. According to our investigation, the value of rural land, village collective interests, and public services in Province J were better developed than those offered in Province H; these factors weakened peasants’ willingness to settle in cities and increased resistance to the urbanization policy. Our interview also shows that more than 900,000 rural residents “moved” from the rural area to the urban area of City S due to land conversion, but only ten thousand of them changed their rural residents’ household in the last three years. At the same time, more than 222,000 rural residents “moved” to an urban center in City WX due to land conversion, but only 11,000 of these residents had actually changed their rural household registration.Footnote 19

Given their strong capacities, local governments in Province J adopted creative policy innovations to meet the central government’s policy objectives and address the needs of local residents, even under high levels of implementing pressure. Differently than in Province H, local governments in Province J have managed to implement novel policies in accordance with Beijing’s goals. Specifically, local governments have adopted various policy measures meant to encourage peasants to settle in cities. Besides, local governments chose to equalize the social welfare provisions between urban and rural residents in their jurisdictions, which minimizes the risk of peasants wanting to keep an urban resident permit, rather than applying for an urban hukou. Such kind of policy innovation takes full account of the peasants’ actual needs and interests. At the same time, local governments in Province J have attempted to harmonize the offer of public services for both citizens with a resident permit as well as for those with a hukou permit; this rendered settling in urban centers more attractive for migrants and thereby increased the urbanization rate. Local policy innovations in Province J demonstrate that there exist creative ways to better alleviate the contradiction between “top-down” goals and “bottom-up” needs. As one cadre told us:

It is not appropriate to regard the urbanization rate of household registration as an assessment index. Improving the basic public services supply is the best way to attract peasants to settle down in cities.Footnote 20

Along with the creative instrument adopted in urbanization rate policy, local governments in Province J also had made big steps forward in the resident permit reform. Specifically, the public services attached to the residence permit policy have been improved to a level that nearly equals the level of services for those with an urban hukou registration. For example, the residence permit in City WX contained 25 items for public service (social security, aid, education, medical treatment, family planning, administrative services, etc.), and there were 24 such items in City S.Footnote 21 In general, local governments in Province J have shown a greater response towards policy piloting, as one cadre noted:

We have fully implemented the residence permit policy, in which the key is to improve the public service contents of the residence permit.Footnote 22

However, it is also evident that the residence permit reform in Province J increases the local financial burden, because the scale of urban migrants in Province J is large, especially in the southern and central parts of the province. Data shows that there were nearly 2.2 million urban migrants in Province J in 2016, while there were 6.47 million urban migrants living in the richest – southern - regions, for instance, cities S, W and C.Footnote 23 Comparatively, there were “only” 450'000 urban migrants in the entire Province H.Footnote 24 Therefore, not all governmental departments positively support the reform. As one local official of the City WX Department of Finance complained:

We need to offer public services in the amount of nearly 120 thousand RMB for a migrant. However, the cost is just for one person. This cost does not take into account the Chinese practice of sending one’s own kids to live with better off relatives and neighbors, who have moved to the city. Thus, per each migrant that moves to a city, we need to consider the possibility that he/she shall bring his/her children, his/her parents and his/her neighbors’, resulting into much higher costs for local authorities. This, because these other individuals, while not working, once arrived in the city, shall nevertheless still enjoy local public services. The expenditure for this group, accounts for more than half of the total local fiscal expenditure.Footnote 25

In sum, even though from a financial point of view these policy innovations are not particularly profitable for local governments in Province J, they are however beneficial to local governance, in safeguarding social stability. Furthermore, although the residence permit policy increases local fiscal expenditures in the short term, in the long run, it contributes to regional industrial development and fiscal revenues, by providing fiscal incentives for the local governments. There are large numbers of migrants in Province J, the majority of whom are part of the industrial labor force, being part of the manufacturing sector and contributing to the increase in financial revenues for local governments. Therefore, the instrument innovations in Province J are sustainable and help to achieve the central piloting goals by applying them to more provinces and localities with similar economic conditions.

Conclusions of the comparative case study

Why do local governments respond so differently to the center’s directives and trial policies without fiscal transfers and resource allocation? The comparative cases analysis shows that local government capacity does matter in determining two kinds of local innovation styles; the main features are shown in Table 3.

Table 3 Comparative case analysis of localities in provinces H and J.
Full size table

We find that localities with relatively higher capacities, for instance, local governments in Province J, are associated with more creative innovation strategies: these local administrations have adequate fiscal resources to offer more public services for peasants and new urbanites and are more forward-thinking in seeing long-term regional development resulting from short-term fiscal expenditures. The logic is that well-developed regions are not always faced with the most urgent problems and have likewise stronger fiscal capacities, higher industrial development, larger scale migration involved in industries, and insufficient quotas of urban construction land needed for urbanization pilots. Local governments, especially in the southern part of Province J, have stronger fiscal capacities to transfer rural land and compensate local peasants to enhance land quotas in support of industrial development. In the long term, it is easier for regions with better economic conditions to keep a balance between fiscal expenditures induced by instrument adoption and local fiscal benefits. For example, the industrial park economy is highly developed in the southern part of Province J, where a large number of FDIs are located; these deliver a considerable amount of local fiscal revenues, helping to push forward the residents’ permit reform.

However, local governments with relatively weaker capacities were unable to find positive and creative ways to align local conditions with central objectives. Instead, they adopted symbolic strategies and, under high levels of implementing pressure, even resorted to distorted policy innovations. China’s political institutions displaying both political centralization and economic decentralization help to explain why localities with weak government capacities are more likely to adopt symbolic or distorted innovation strategies. Such localities prioritize central directives over citizens’ needs (Ahlers and Schubert, 2022) to maximize their fiscal and political interests, often leading to collusive behavior among local governments (Zhou, 2010), exacerbated by the problem of information asymmetry within China’s bureaucracy (Liu et al. 2024).

In general, it can be summarized that local governments in Province J are capable of reaching a balance between the “top-down” piloting goals and “bottom-up” peasants and migrants’ demands. On the contrary, local governments in Province H, like the one of City S, chose to seek to accomplish centrally-driven urbanization rates through administrative reorganization, although this process may harm local governance.

Further discussion: the costs of policy piloting

The comparative cases from nine localities in China’s provinces H and J reveal the differing ways local governments respond to central directives and the underlying political-economic dynamics. Generally, most local governments in Province J adopted creative innovation strategies that significantly improved local governance. However, in some cases, local governments implemented symbolic or distorted innovation strategies, which only slightly improved or even harmed local governance. Therefore, the costs of experimental governance under a hierarchical system arise from local governments’ adoption of symbolic or distorted policy instruments.

When discussing the costs of policy piloting, we must examine the cost-benefit structure of a national pilot program. Specifically in this study, the central government is more likely to benefit from national piloting by selecting effective policy instruments from localities while avoiding the provision of fiscal transfers and resources. However, local governments may bear both the benefits and costs of participating in a national piloting program: when fulfilling central piloting goals improves local governance, the benefits are prominent; when fulfilling these goals comes at the expense of local needs, the costs outweigh the benefits.

As a result, the costs of policy piloting are connected to the large-scale instruments innovation across localities with different local conditions in China’s bureaucracy. It is evident that policy piloting under hierarchy represents a possibility to gain advantages for many local governments; only a few among them, however, are capable of doing so. As the NUP is sort of a top-down mandatory policy piloting, nearly half of the local governments are assigned piloting tasks. Eventually, though, the central government only selects a limited number of policy instruments from the localities, which are suitable to be implemented nationwide (as we described before), leaving the rest with virtually no gains and lots of wasteful resources devoted to an unusable plan.

In fact, a related problem is who should ultimately bear the costs for the majority of the unsuccessful policy instruments tested at the local levels. A further problem is that there is no guarantee that even those so-called “successful” and “effective” policy instruments tried at the localities can successfully be applied by other local governments in other parts of China. In fact, this seems very unlikely, due to the difference and variations we have found in the cases. Taking into consideration the huge differences regarding geographical location and economic conditions across localities in China, the logical conclusion is that for many local governments the national piloting goals will not be easy to reach. Nevertheless, local governments need to comply, at least formally, with centrally mandated plans, even if these turn out to be detrimental to the local conditions. As we can see from our case studies, administrative reorganization reforms are chosen to comply (even though symbolically so) with the center’s goals towards policy innovation.

On the national scale, the phenomenon of “fudging statistical data” observed in Province H, has been observed in other localities across China. For instance, the urbanization rate regarding the population with an urban hukou in Province A increased rapidly: from 22.7% in 2014 to 31% in 2017;Footnote 26 during this time which process the “county-district conversions” land reform process, took place in City TL, starting in 2015. In 2018, a new administrative reorganization reform has taken place in City WH, Province A. As explained throughout the paper, these are not real migrants who move and obtain an urban hukou, but it is merely the land where they live which gets classified as “urban” overnight, so that its residents become “urbanites”.

The above discussion highlights the need to distinguish the benefits and costs borne by different levels of government and actors in policy pilots in order to better evaluate the cost-benefit structure of such programs. A policy design perspective should be incorporated into the management of national policy piloting, focusing on identifying the factors that shape local innovators’ differentiated response strategies to central piloting goals and subsequently optimizing these goals.

Concluding remarks

This article explores the impact of local conditions, particularly local government capacity, on local governments’ response strategies to central piloting goals in the absence of fiscal transfers and resource allocation. We find that local governments may adopt differentiated strategies—ranging from creative to symbolic to distorted innovations—depending on whether central goals are accompanied by quantifiable indicators. When local government capacity is strong, governments are more likely to adopt creative innovations to reconcile central piloting goals with the interests of local citizens. However, when local government capacity is weak, governments are more likely to adopt symbolic innovations under low implementing pressure (with vague goals) and distorted innovations under high implementing pressure (with quantifiable goals).

These findings contribute to the literature on local governments’ response strategies to central piloting goals. First, this research broadens the understanding of how local governments respond to central piloting goals by revealing a range of strategies, including creative, symbolic, and distorted policy innovations, thereby shedding light on the dynamics of central-local relations in experimental governance from the perspective of local governments (Ettelt et al. 2022; Zhu and Zhao 2021). Second, it also highlights the political-economic dynamics of local governments’ responses, particularly the role of local government capacity, a factor that has been largely underestimated in the literature (Guo et al. 2021; Ma and Liu 2024).

In addition, we also argue that policy piloting at the localities in large-scale piloting programs, under a hierarchal system, is not necessarily conducive to improved local governance. Only few of the local pilots can ultimately be adopted by the central government as instruments nationwide; in some cases, local policy pilots are carried out to fulfill the central goals but do not meet local needs. Such kinds of symbolic and distorted innovation strategies are rooted in China’s experimental system, where local cadres treasure the center’s goals as the priority. As a result, social costs are generated along with some benefits for experimental governance after several rounds of trial-and-error schemes.

These findings help us deduce implications for better design of policy piloting in China’s bureaucracy, where local officials are not elected by voters but appointed by upper-level officials and local governments’ policy innovation is first and foremost tasked with meeting the goals of the center. First, it is crucial to evaluate and enhance local governments’ capacities in implementing national piloting goals, while setting diverse yet achievable goals for localities to maintain an appropriate level of implementing pressure. Second, the central government should not only adjust its evaluation indicators (especially for quantifiable goals) by incorporating feedback from localities but also provide policy guidance to local implementers and innovators throughout the piloting process.

Finally, this study has several potential limitations. First, it did not consider all social actors (such as enterprises, communities, and others) when discussing the social costs of national piloting programs, which limits a deeper investigation into the cost structure of these programs. Second, the external validity of the findings is limited to provinces J and H, as these two cases cannot represent all variations across localities in a large and diverse country like China. Third, the two key variables—local government capacity and implementing pressure—were measured qualitatively, which highlights the need for cross-sectional panel data studies to better quantify the driving factors behind local governments’ differentiated response strategies to central piloting goals.