Abstract
To explain the poor results of several international collaborative projects dedicated to exploring eco-city concepts in China, some scholars pointed out that Chinese policymakers operate a form of cherry-picking that prefers technical knowledge to policy proposals. This approach is consistent with the hypothesis of a presumed Chinese style of policy transfers, according to which Chinese policymakers learn from abroad selectively and gradually. This study aims to test this hypothesis by analyzing one of these eco-city collaborations: the transfer of “Careful Urban Renewal” to the city of Yangzhou. Through a longitudinal study that looked at policy developments over the period 2003–2019 and a focus on the “demand side” of policy mobilities, it illustrates that transfers concerned several policy aspects, going beyond technical recommendations. This disconfirms the hypothesis of a consistent policy transfer style, as no specific policy preference could be spotted. Rather, the analysis of the policy process illustrated that different policy transfer styles emerged in the period considered. Variations in policy transfer style can be ascribed to the willingness, motivations, and interests of specific individuals, in particular city party secretaries and local bureaucrats, as well as to the room of maneuver and resources these actors possess, in turn, determined by the institutional structures in which they operate. Situational factors also led to variations. While this single case study cannot encompass all variations of policy transfer style, it is nonetheless hoped that this research inspires more longitudinal studies, which in turn can contribute to a promising discussion in policy studies and urban studies.
Introduction
Since the early 2000s, several Chinese cities have participated in international initiatives to explore sustainable approaches for urban development and exchange knowledge in the field of urban planning. Two famous examples are the Sino-British and Sino-Singaporean collaborations for the conceptualization, planning, and construction of eco-cities in the outskirts, respectively, of Shanghai and Tianjin. Other initiatives focused on providing technological assistance, planning blueprints, or passing on knowledge in planning and architectural fields (De Jong et al., 2016; Fan, 2014; Pow and Neo, 2013; Zhou, 2021). The modes of cooperation and the objects transferred were therefore diverse, ranging from technical solutions to urban planning practices and policy ideas (Chang, 2017). Urban studies scholars analysed some of these collaborations, focusing mostly on cases of eco-city construction (e.g. May, 2008; Chang and Sheppard, 2013; De Jong, 2013; Joss and Molella, 2013; Fan, 2014; Hult, 2015; Caprotti et al., 2015). They illustrated that these projects were not very successful, as their impacts on planning and construction practices were limited, while the solutions adopted were inappropriate for the problems targeted.
Acknowledging the “trans-local knowledge sharing” nature of these initiatives (Chang, 2017), De Jong (2013) attributed these outcomes to a distinctive Chinese style of policy transfer, according to which Chinese policymakers import foreign knowledge selectively and gradually, which leads to partial or incomplete transfers. The hypothesis of a policy transfer style, according to which it is possible to find recurring patterns in the way policy transfers occur, is very stimulating, as it joins a nascent discussion on the impacts of China’s specific politics, institutions, and traditions on its policymaking characteristics (cf. Heilmann and Perry, 2011; Qian, 2018, 2021; Mei, 2020). This discussion is not only interesting for policy analysts but can also contribute to debates in urban studies dedicated to cities ‘elsewhere’ and their potential to “set the terms of discussion” for comparative research and further theoretical development (Ren, 2022: 1743; Zhou, 2021). Moreover, it can contribute to the discussion on the somewhat less-explored “demand side” of policy mobilities, which requires a deeper examination of the agency and practice of actors at the receiving end of the mobility process (Silvestre and Jajamovic, 2021: 2313) and of “how policy is localized” (Temenos and Baker, 2015: 842). In particular, the concept of policy style lends itself well to understanding “the relationship between politics and policy” (Howlett and Tosun, 2018: 4) and the impact of formal and informal institutions on policymaking processes. This focus can contribute towards identifying those “shared features rooted in institutional commonality” (Ren, 2023: 2) that would make studies of urban China more relevant for comparisons and theorization. Moreover, it builds “a greater dialogue with policy studies” to produce more nuanced analyses of the “receiving end” of policy mobilities (Silvestre and Jajamovich, 2021: 2313).
Interestingly, no follow-up studies have validated the claims of a distinctive policy transfer style and identified its characteristics – if any. As also illustrated by Chang (2017), there are not many analyses of collaborative projects from a policy mobility (or a policy transfer) perspective dedicated to China. Moreover, these have not necessarily addressed the specificities of these phenomena in China. 1 In particular, the claim that policy transfers in China occur selectively and gradually shall be validated through observing cases of transfer longitudinally. But with few exceptions, a consideration of the time dimension is missing in existing studies. The evaluation of policy transfers and their outcomes is mostly based on analyses conducted on the spot, realized during the cooperation project, or after its conclusion. This fails to capture possible policy developments, which take more time to manifest (Colomb, 2007; Miao, 2018) and also fails to identify the presence of a policy style.
In order to address these questions, the paper analyses the case of the transfer of “Careful Urban Renewal” (hereinafter ‘CUR’) to Yangzhou, a city situated in Jiangsu province. Originating in West Berlin in the 1970s, CUR broke with the mainstream practice of urban redevelopment centred on large-scale demolitions. Architects and planners introduced new ideas to avoid the relocation of residents, preserve the existing urban layout, buildings, and social structures, and reduce environmental impacts (Bernt, 2003). In the early 2000s, CUR was promoted to Yangzhou via a project run by the German international cooperation agency GTZ. Established under the name of “Eco-city planning and management program”, the project covered a variety of topics related to eco-city planning, ranging from clean transportation to water management. CUR was included in this package of proposals and targeted the city’s old districts (the ‘Old City’). The inquiry, realized six and then eleven years after its completion, revealed that the local authorities in Yangzhou took up many of the recommendations made by GTZ, experimented with different ideas, introduced new practices, and started new renewal projects. Therefore, the case lends itself well to identifying the long-term impacts of the policy transfer, the characteristics of the transfer process, and which factors mattered in the process.
The analysis is based on a longitudinal case study that monitored policy developments over the 2003–2019 period. The inquiry was conducted between 2013 and 2015 and later complemented by new field visits in 2018 and 2019. It focused, first, on how foreign experiences were communicated, reconstructing the activities that took place during the international cooperation project. Second, it investigated the local policy learning process and knowledge use and identified policy impacts. A combination of research methods was used. These included 61 semi-structured interviews of variable length with officials and residents in Yangzhou, with GTZ agents, and Chinese and foreign experts located in China and Germany. 2 The inquiry was also supported by several on-site observations, and a thorough analysis of policy documents, urban plans, and various reports produced by GTZ. Secondary literature and newspaper articles provided further information about the case. Furthermore, the consultation of bibliographic references in Berlin and interviews with two experts helped retrace the origins of CUR. This methodological choice aimed at “following policies” while focusing on policymakers’ and other agents’ practices, allowing a careful assessment of policy impacts (Colomb, 2007; McCann and Ward, 2012).
The study reveals that policy change occurred at different levels in Yangzhou, from policy instruments to the ideas underpinning planning policies. These changes drew significantly from the proposals made by GTZ. Policy adoption was also selective and gradual. However, when broadening the time perspective for observations and focusing on local policymakers’ adoption and localization activities, one can notice that these characteristics are not attributable to a specific and persistent Chinese style of policy transfer. Rather, policy styles changed considerably over time, shifting from a phase in which the policy preferences were substantially conserved to a phase in which policy reforms were bolder and moved faster, to later shift again towards a more gradual process of policy development. This suggests the presence of several styles. These variations can be ascribed to the willingness, motivations, and interests of specific individuals, in particular city party secretaries and local bureaucrats, as well as to the room for manoeuvre and resources these actors possess, in turn, determined by the institutional structures in which they operate. Moreover, various situational factors also impacted the mode of policymaking. Therefore, it is not possible to talk about a single Chinese policy transfer style, but rather a multiplicity of styles that are manifested according to variable configurations of these different factors.
To demonstrate these points the article proceeds as follows. The next section introduces some contextual information about Chinese urban politics and planning, focusing on the aspect of urban renewal. This makes it possible to understand the situation before the policy transfer, in order to later identify at which level policy change occurred. It also presents a review of the literature on the transfer of best practices in urban planning in China, introducing and discussing the hypothesis of a Chinese-specific policy transfer style. After introducing the research framework, the paper provides an analysis of the case study. The concluding section summarizes the findings and suggests future research directions.
Urban renewal, the transfer of best practices and policy styles in China
CUR was proposed to Yangzhou authorities to replace a practice of redevelopment that emerged in the 1990s when China transitioned from a centrally planned economy to a market-oriented economy. China’s economic transformations indeed had important consequences for urban planning and city development. Many competencies were shifted from the central government to cities, which became the main responsible units for local development (Yeh and Wu, 1999). Together with this decentralization, the central government also established a series of targets to evaluate local governments’ performances, and more specifically local party cadres (Edin, 2003). Called the “cadre responsibility system”, for a long time this set GDP or fiscal revenue growth as a major criterion for evaluation (Li and Zhou, 2005). This constituted a powerful driver for urban development. Combined with a series of crucial reforms, initiated in the 1980s, that concerned fiscality, land, and housing policies, these policy changes made urban construction and redevelopment key solutions for local governments to boost GDP growth, while giving party cadres more chances to be promoted (Yeh and Wu, 1999; Wu et al., 2006).
Local governments thus began to intervene in urban spaces and profoundly transformed the aspects, functions, social composition, and wealth of Chinese cities. Many old neighbourhoods were demolished to make space for modern road infrastructures and high-rise buildings (Xu et al., 2009). These were instrumental to boost local revenues and increase land value (Hsing, 2010). Some neighbourhoods were gentrified, while others changed their functions (Yang and Chang, 2007). The historical fabric was destroyed “at an unprecedented pace” (Whitehand and Gu, 2007: 650), and many residents were displaced. When preserved, old neighbourhoods were sometimes transformed into tourist attractions, following an approach of “heritage commodification” and “symbolic urban preservation” which considered old cities only as sources of revenue (Su, 2010, 2015; Zhang, 2008).
Acting as entrepreneurial states, local governments either developed policies to attract investors (Wu et al., 2006), and in this way promoted urban redevelopment with the help of the private sector (He and Wu, 2005), or even led the entire process, under what Wu (2015) calls “state-dominated urban redevelopment”. In the latter case, the agents who conducted urban redevelopment were agents of the local state, such as district governments and state-owned enterprises (SOEs). These played a central role, as they were given easy access to land use rights and credits from state-owned banks, as well as easy conditions to acquire properties in redevelopment areas. They were responsible for demolition and land clearance, could act as landlords, and could also control the newly developed properties as part of their investment portfolio (Wu, 2015: 655). Local governments could thus employ their many resources to redevelop land, and this in turn helped them boost their revenues and their competitiveness (He and Wu, 2009; Hsing, 2010).
The project carried out by GTZ thus had the ambitious goal of inspiring policy change and breaking with these practices, promoting best practices for the conservation of the Old City, its functions, uses, and social structure. Interestingly, this case was singled out as a “notable example” of eco-city construction, in which “the importance of local public participation and conserving the Old City fabric” was emphasized (Pow and Neo, 2013: 2272). However, a study published in 2014 noted that although international assistance had demonstrated the validity of its proposals to reform the local practices of urban renewal, the local government kept its old model of urban renewal. “Technical expertise was easily introduced”, but the project “did not influence a governmental shift in policy priorities” (Fan, 2014: 658). These findings resonate with those of other studies, confirming that international assistance can do little or nothing to change local urban planning and conservation practices (Su, 2010) and that only a few recommendations are adopted (De Jong, 2013; Hult, 2015).
De Jong (2013: 98) suggested that this partial adoption of foreign knowledge can be explained by China’s particular style of policy transfer. This is an interesting argument, assuming the presence of country-specific and sustained patterns of policy adoption, which persist over time (see also Zhou, 2021). The concept of policy style was developed in public policy studies to describe and explain the propensity of policymakers to deal with public issues in recurring patterns. These, in turn, can be ascribed to specific “historical legacies and institutional structures that routinise decision making” and are at the heart of “distinct implementation logics” (Mukherjee and Howlett, 2016: 26), as well as of specific policy preferences, with certain types of instruments being preferred over others (Howlett and Tosun, 2018).
Applying this concept to evaluate policy transfers in China, De Jong (2013) argues that the latter are characterized by selectiveness and gradualism. We can observe “gradual policy change and eclecticism in the adoption process”, which conform to “domestically dominant institutional practices and preferences” (2013: 99). These features explain the existing implementations of eco-city concepts. Namely, local policymakers in China operate a form of cherry-picking in which technical advice is preferred to other transfer objects, such as procedural or organizational proposals. In the case of the collaborative project analysed by de Jong, the local government only implemented a few initiatives, but did not follow up with the “integrated future vision” that the experts had provided (2013: 98). This, he argues, can be ascribed to a “path-dependent view of institutional transplantation” (2013: 91), in which Chinese policymakers import foreign concepts based on “national self-interest” and following the above-mentioned patterns. He also observes that the local government ended up making “considerations of rapid construction, limited budgets, attraction of prestigious companies and promotion of the area as a Low Carbon City” principally for branding purposes (2013: 98), as observed elsewhere in China. Suggestions to reform administrative practices or to experiment with citizen participation were discarded, as they were deemed disruptive to the existing administrative arrangements, and incompatible with political and administrative traditions. Moreover, he observes that interdepartmental coordination “to achieve an integrated implementation plan” never happened, due to coordination challenges in the Chinese administration (2013: 98).
De Jong formulates this hypothesis based on his consultancy experience in the collaborative project and on the insights of literature dedicated to foreign-inspired policy reforms introduced in China in the 1980s and 1990s. However, the evaluation of the impacts of the project and the characteristics of the transfer process occurred close to its conclusion, i.e. too soon. An extension of the time frame would have generated more insights and helped spot more “policy changes triggered (or informed) by learning from abroad” (Dussauge-Laguna, 2012: 575–6). Such a time extension is also useful to discuss the presence of a specific policy style. The concept of policy style indeed assumes that “the same actors, institutions, instruments, and governing ideas dominate policy-making for extended periods of time, infusing policy with both a consistent content and a set of typical policy processes or procedures and actors” (Howlett and Tosun, 2018: 5). As such, policymakers are assumed to only import policies that are compatible with their policy frames and preferences. Moreover, the policy style is thought to persist over time. Therefore, in an analysis interested in identifying policy styles the dimension of time must not be missing, as conjunctural shocks and cyclical developments may impact the patterns of policymaking (Freeman, 1985).
Another aspect to underline is that De Jong (2013) evaluates the impacts of the policy transfer from his perspective, while that of the recipients of the transfer process is missing. The result is an analysis that implies a “relative passivity and stability of interests” of local actors (Silvestre and Jajamovich, 2021: 2312). De Jong (2013) talks about “national self-interest”, assuming a homogeneity of interests among policymakers in China. Furthermore, agency is almost absent in this characterization of policy transfer style, which is focused above all on the outcomes and practices of transfer. De Jong (2013) speaks of generic “Chinese policy-makers”, but this label overlooks the differences in interests and motivations that can be found within any government in China, from the central level to the municipal or county level. As pointed out by Qian (2018) in a paper discussing China’s policy styles, government departments have their own interests, and this aspect explains the lack of policy coordination also observed by De Jong (2013). For this reason, examining the methods, motives, and individuals involved in using knowledge is also essential (Silvestre and Jajamovich, 2022: 1456). This would illustrate, for instance, that notwithstanding the absence of interdepartmental coordination, Chinese bureaucracies can also establish instruments to enhance policy coordination (Qian, 2018: 203). This suggests that policy styles can also change over time because of different impulses. Moreover, a change in policy styles can be triggered by “the slower and less direct diffusion of administrative ideas from one tradition to another” (Howlett and Tosun, 2018: 7), an aspect that makes it even more compelling to carry out a longitudinal analysis of the impacts of policy transfers, of local appropriation processes and the agency of local actors.
Here some clarifications are nonetheless needed. First, this study does not make any substantial distinction between the transfer process and the process of policy development, as suggested by Stone (2017). Drawing from policy mobilities and policy translation studies, Stone proposed to understand policy transfers as evolving processes of “assemblage”, “bricolage” and “localization”. Transfers can be seen as open-ended processes of trial and error in which learning, translation, and interpretation continue to occur. Such conceptualization implies moving away from the binary distinction between success and failure, as well as from the qualifications of “uninformed”, “incomplete”, and “inappropriate” transfers much used in the policy transfers literature (Chang, 2017). Rather, we shall understand policy transfers as long-term, intergenerational processes of policy development that unfold and evolve across time and may trigger unexpected results (Dussauge-Laguna, 2012; Miao, 2018; Stone, 2017).
Second, identifying a single Chinese style of policymaking is a daunting task, as one would have to analyse policy developments in different policy sectors and make systematic comparisons. This can only be achieved after a collective and cumulative effort. Nevertheless, this exercise could prove to be of little use because there is marked variability among policy sectors (Freeman, 1985). For instance, Qian (2018) illustrates that policy styles in China may present different characteristics depending on the sector of intervention (e.g. social policy or economic policy). In some cases, they are more top-down, “reactive and imposition based”, with local bureaucracies interpreting and implementing the wishes of the central government. In others, they are more proactive, “responsive to public complaints” and use local knowledge (Qian, 2018: 203). This suggests the usefulness of limiting the discussion about policy styles to specific policy sectors, and it is in this sense that the contribution of this article must be understood.
The research framework
A focus on policy styles requires an assessment of how existing administrative and political institutions affect policy processes and their outcomes (Howlett and Tosun, 2018). This in turn makes it necessary to adopt appropriate units of analysis. Howlett and Tosun (2018: 8-9) suggest focusing on both policy agents and the structures within which they operate, as it is assumed that the former act “within an institutional context that at least in part determines their behavior”. In operational terms, on the one hand, this makes it necessary to identify the relevant actors in the policy transfer process and their motivations. On the other hand, it requires an analysis of the factors restricting and/or facilitating the transfer process, and in particular the role of institutional constraints, which are central in the identification and evaluation of policy styles (Howlett and Tosun, 2018).
As for policy outcomes, Howlett and Tosun’s discussion does not dwell on this aspect, but they also represent a component of policy styles, as these are associated with specific policy frames and preferences, i.e. policy ideas and instruments (Howlett, 1991; Mukherjee and Howlett, 2016). Within the framework of this study, we are particularly concerned with policy transfer outcomes. To this end, the taxonomy proposed by Howlett and Cashore (2009) to identify levels of policy change can be useful, particularly if the transfer object is not a policy ‘off the shelf’, as in the case of legislative bills accompanying specific policy programs (Dolowitz and Marsh, 1996), but rather a set of recommendations and guidelines (Vettoretto, 2009). In this case, to assess transfer outcomes, we shall specify how this set of recommendations impacted on various policy components. If policy frames and policy preferences remain constant after the policy transfer, i.e. policy ideas and policy instruments remain the same, then we can talk about policy style consistency. If these change, then the policy style has changed.
Re-elaboration of Howlett and Cashore’s (2009) Taxonomy of Policy Components.
Finally, as mentioned above, a consideration of the time dimension is needed to both evaluate the impacts of policy transfers and observe how government or regime changes and other “situational factors” (e.g. conjunctural shocks) can lead to an alteration of policy style (Freeman, 1985: 477-9). To consider these aspects in the analysis, the suggestions made by Dussauge-Laguna (2012: 581) to focus on political cycles “and the ‘windows of opportunity’ that come with them” to integrate time in the analysis of policy transfers is, in this respect, particularly useful. These elements can be systematically considered together with conjunctural shocks and cyclical developments (Freeman, 1985), which can also lead to radical changes of policy style.
Following this framework, the analysis will be organized as follows. The period 2003-2019 will be considered, and in turn, divided into two phases. The first runs between 2003 and 2008, corresponding to the years of international cooperation; the second runs between 2009 and 2019, recording policy transfer impacts ex-post in the medium to long term (see Colomb, 2007). For each phase, the policy transfer outcomes are first examined. Second, the analysis will dwell on the relevant actors of the policy process and on the factors that impacted the transfer process. A discussion section finally concludes each phase, evaluating the argument of a policy transfer style. For each phase, a brief description of the events will be also provided, to help situate policy developments.
2003-2008: Selectiveness and preservation of policy preferences
In 2003, Yangzhou was selected by the State Environmental Protection Agency, as it was then known (SEPA, now the Ministry of Ecology and Environment – MEE) as a national pilot for eco-city construction. In the same year, the city government began collaborating with GTZ to explore the model of CUR. This can be understood as a model of city planning and urban renewal. In Berlin, its introduction marked a passage from a top-down paradigm of planning to a more socially oriented, environmental-friendly, inclusive, and participatory paradigm, careful to preserve the existing city and promote its mixed-use (Colomb, 2012).
However, what travelled to Yangzhou is not really CUR, as it was established in Berlin in the 1980s. To do so, GTZ would have had to promote the transfer of the complex set of laws, regulations, standards, and social practices that supported its implementation. Rather, as in many other cases of policy mobility (Montero, 2020), what it promoted was knowledge about CUR, leveraging the experience of Berlin while combining it with other ideas that circulate internationally and promote change in the policy, organization, and practices of urban renewal. These considered different aspects, such as the improvement of housing conditions and infrastructures, the provision of open space, local economic development, and the conservation and management of heritage (Zhu L. et al., 2007). Institutional responsibilities, planning and management solutions, financing requirements, and implementation schedules were also included.
Retaining only technical suggestions
Between 2003 and 2007, GTZ agents used a large variety of activities to promote CUR. Among them, we find study visits to Berlin and to other cities in Germany and China, and the invitation of experts from different organizations (international, nongovernmental, private) to provide expertise on various policy sectors. All these activities were accompanied by a training program for local officials, the production of studies, reports, and recommendations on a variety of policy sectors, workshops with different members of the city government, and the realization of pilot projects in three sites. The project demonstrated to the local administration that it was possible to conduct urban renewal in another way. However, this was not enough to inspire policy change, as captured by the following interview abstract: What they [the administration in Yangzhou] were quickly adopting were all the technical aspects of the project, this was very easy to adopt, so what to do and how to repair [buildings] and how to bring in the infrastructure and how to improve the situation for the people living in those areas. What is much more difficult and where the project failed […] was to import certain structural changes, and also some organizational patterns that would help […] establish a process of on-going conservation and rehabilitation (GTZ Expert Berlin, 2014, personal communication).
Other interviews with GTZ experts confirmed this assessment. Further demonstrating the lack of policy change, shortly after the conclusion of the pilot projects, the city government carried out a new redevelopment project, based on demolition and resident relocation. The party secretary pushed for its realization, and it was implemented by one of the local SOEs. Thus, the policy preferences of the local government remained the same, while the policy transfer did not have any substantial impact at any policy level identified by Howlett and Cashore (2009).
Actors, institutions and political cycles
A combination of factors can explain these outcomes, but local actors, their motivations, their political resources, and the local institutional set-up stand out above all. Two main constellations of actors mattered in the international collaboration: a group pro-CUR, represented by a few local officials employed in the urban planning department, who collaborated with GTZ and were supported by the mayor; and a group composed by the local party secretary and other local officials employed in the urban construction department, in favour of redevelopment. The first group can be classified, together with GTZ experts, as the ‘transfer agents’, as through their cooperation they carried out the pilot projects and discussed many policy ideas. However, this group did not have enough political resources to push for the adoption of a new urban renewal approach, since the ‘pro-redevelopment group’ was more powerful.
Its power is derived from the institutional set-up of Chinese cities. These have two major political leaders: the mayor and the local party secretary. It is customary that a party cadre, after having served as mayor in a city, becomes its next local party secretary, covering a position of greater prestige and power within the local government. The party secretary can select departmental directors, impose administrative reshuffling, impose his/her decisions, and has more power than the mayor. 3 In the years of GTZ collaboration, Ji Jianye was first mayor and then party secretary. Nicknamed the “bulldozer mayor” by the press (Lü and Ju, 2013), he had certain policy frames and preferences that in that period also predominated among many other city governments (see above). These consisted of the demolition of old buildings, the relocation of residents, and the change of space functions as the best approach to solve the problem of urban decay and improve the aesthetics of old neighbourhoods. Therefore, demolition and reconstruction were preferable to conservation. The latter would have also cost too much time, a resource that party cadres do not have. Ji’s approach was also dictated by the imperative of boosting GDP growth during his mandate, which would have guaranteed him better promotion chances. As illustrated earlier, if this was his motivation, we must not however forget the structures in which party cadres operate, which encourage transformations of urban spaces as a means of fostering GDP growth and maximizing promotion chances (Wu F. et al., 2006).
A selective policy transfer style
When looking at the outcomes of policy transfer in this specific phase, many of the observations made by De Jong (2013) and other scholars (Fan, 2014; Hult, 2015) can be confirmed. At the end of the project, the transfer concerned exclusively technical recommendations. Thus, the aspect of selectivity is confirmed – only technical suggestions were retained, while proposals to change the governance of urban renewal were not adopted. The study also confirms that entrepreneurial considerations such as those described by De Jong (2013) and various urban studies scholars (Hsing, 2010; Wu, 2015), i.e. using the transformation of urban spaces as a means to foster GDP growth, dominated policy-making and illustrated the persistence of specific policy frames and policy preferences. Different understandings of urban planning and the goal of boosting GDP growth thus made GTZ proposals incompatible with local dominant policy frames and preferences.
Taking up the concept of policy styles proposed by Howlett and Tosun (2018), here it is possible to find not only the same governing ideas, but also the same actors, institutions, and instruments. The redevelopment project carried out shortly after the conclusion of the collaboration with GTZ featured the same constellation of actors that Wu (2015) describes when he writes about “state-dominated urban redevelopment”. Even the instruments used were the same: the redevelopment project was financed through operating via the SOEs, which can easily obtain credit from state-owned banks. Moreover, it was observed that local political leaders work within a precise structure of constraints that derives from the institutional set-up, in particular from the rules regulating the promotion of party cadres and the incentive structure created by the central government. Hence, these findings confirm the presence of a selective (and presumably gradual) policy transfer style, in which local policymakers import ideas only according to their interests.
However, shifting the focus to local government actors reveals a slightly more complex situation than the one illustrated by De Jong (2013). In fact, within the category of ‘policymakers’, both referring to administrative and political personnel, we can find different policy preferences and interests. One group of administrators and the mayor did not share the same vision and policy preferences as the party secretary and another part of the administration. However, their chances of promoting policy change were limited as their power resources were limited. But this situation has not remained unchanged, as illustrated in the next section.
2009-2019: A radical change of policy style?
Between 2009 and 2013, the city government adopted a series of plans and policy documents that translated the proposals made by GTZ a few years earlier. Moreover, it carried out some renewal projects following the GTZ model. Therefore, contrary to expectations, GTZ’s contribution proved decisive in fostering a change in the frames, policies, and practices of urban renewal in Yangzhou. In the following years, however, this reform momentum slowed down. Between 2014 and 2019 the development of measures that followed GTZ recommendations became more sporadic, while some were never adopted. Furthermore, the risk emerged of a return to the old urban renewal model. This makes it more difficult to identify a consistent policy style for the policy transfer process.
Adoption of many (but not all) GTZ policy proposals and a partial change of policy preferences
In July 2009, the city government established an Old City Office (OCO). Close to what GTZ recommended, it was conceived as a coordination and “implementing agency for the protection of the old city”, and it was “committed to all its aspects”, ranging from the establishment of policies and measures for old city protection to their implementation (Yangzhou Construction Bureau, 2010). Between 2009 and 2013, the OCO prepared new plans and policies, following many GTZ recommendations. Given their large number it is not possible to indicate in a single article what was adopted, but a few examples will be provided. Among them, the “Opinions for the Renovation of Yangzhou Old City Traditional Houses” (henceforth the ‘Opinions’), issued in June 2011, reveal that the OCO fully endorsed the policy, principles, and measures proposed by GTZ. This made the establishment of regulations and standards for private housing renewal a main priority of old city protection.
Regulations were intended to target residents’ needs and promote resident participation (Zhu et al., 2007: 44-45). This participation was planned to not only benefit residents, but also free the local government from considerable financial burdens, since redevelopment operations required an enormous effort from the state coffers. It is upon these bases that the OCO prepared the Opinions, which established the basic principles and measures for the renovation of private houses. To encourage resident participation, the Opinions also established special subsidies. These were designed following a scheme prepared by GTZ for the realization of the pilot projects. To understand the importance of this document, one must also look at the new master plan introduced in those years. This revealed substantial changes in urban renewal goals compared to that of the early 2000s. This established that the Old City had to change functions, reducing residential land use through relocating residents. This indication disappeared from the new edition (2012-2020), which called for the preservation of the existing residential land-use ratio and the gradual upgrading of the residential environment.
Policy Changes in Yangzhou.
However, not all the recommendations made by GTZ found a corresponding translation in Yangzhou. An interesting example concerns the introduction of measures to promote the upgrading of publicly owned residential buildings. Public housing represents roughly half of the buildings in the Old City. This situation is problematic for its upgrading. In some cases, no owner can pay for the renovation, as the SOEs to which they belonged closed. As for municipal housing, the agencies in charge claimed that rents are insufficient to fund repair and maintenance and that the city government’s annual budget can only cover part of these costs. For these reasons, GTZ proposed to privatize public housing. In 2012, the OCO issued a tentative policy following this idea, extended the subsidy program to public housing, and applied it in a pilot area. However, no family applied for it, leading the OCO to resume policy design work. But at least up to 2019, no new measures were established.
Interestingly, between 2014 and 2019 only a major policy achievement can be registered: the preparation of the “Yangzhou Old City Conservation Regulations”, issued in 2016. Confirming the objectives and instruments of the urban renewal policy established a few years earlier, these gave greater legal force to the new policy of urban renewal, which until then was only expressed in less binding administrative documents. Indeed, the publication of the old city conservation plan had not prevented the new mayor, who arrived in 2014, from announcing two major redevelopment projects in the old city. Eventually, these were never realized for reasons that will be explained shortly. However, they illustrated that the risk of a U-turn in urban renewal was very high. The observation of policy outcomes however illustrated that policy frames and preferences have not remained unchanged. On the contrary, the changes at different levels of the urban renewal policy suggest the presence of another policy style.
Actors, institutions, political cycles, and situational factors
Once again, several factors explain these outcomes. The years 2009–2013 and 2014–2019 corresponded to new political cycles for the city of Yangzhou, in turn related to changes of party secretary. The first change (2009–2013) opened a policy window for the group in favour of CUR. Wang Yanwen, mayor at the time of the GTZ project and new party secretary in this phase, promoted the founding of the OCO and put one of the officials at its head who closely collaborated with GTZ and continued operating as a transfer agent. Backed by this political support, which provided the OCO with several resources, this new office made several policy proposals, adopted various policies, and began new renewal projects.
Chances to promote a new urban renewal policy also resulted from the combination of other factors. The redevelopment project wanted by the previous party secretary received many criticisms from the two local assemblies, the People’s Congress and the People’s Consultative Congress, as well as within and outside the administration. Financial constraints also prompted a revision of the urban renewal model. Starting in 2008, the central government introduced some measures to slow down urban transformations, reducing the possibility of financing various projects by forcing state-owned banks to toughen their conditions for credit provision (Keith et al., 2013). This prompted a rethink of renewal projects since it was no longer possible to rely on the usual model of financing. An increase in land prices between 2008 and 2010 also made the costs of relocations more exorbitant, prompting the administration to look for less expensive solutions. The model proposed by GTZ therefore began to convince more officials in the administration, since it envisaged a limited use of public money. This favoured a rapprochement of policy ideas between different members of the administration, which found in CUR a way to answer their specific concerns.
However, this favourable phase for policy development partially closed with a new political cycle. In 2014, a new party secretary and a new mayor came to power. The party secretary was less interested in following the predecessor’s policy and withdrew his support from the OCO. Without resources, this office could no longer lead policy initiatives, and this led de facto to a closure of the policy window. For instance, it was no longer possible to rethink the public housing policy, as this was strongly opposed by the responsible agencies. The party secretary also decided to shift many of the OCO’s responsibilities to the control of the city districts, which then became project initiators in the Old City. Without the support of the party secretary, it was also more difficult for the OCO to question the new mayor’s redevelopment projects. What eventually stopped their realization was residents’ protests and the concomitant pressure from the central and provincial governments, which became less tolerant of such protests and sent warnings to the Yangzhou authorities. The central government also began sending signals of change concerning urban redevelopment. In a series of public speeches, President Xi Jinping himself criticized the indiscriminate development of cities and the irreparable loss of urban heritage linked to redevelopment. Therefore, the pressure coming from the central government prevented a policy U-turn.
It was then once again an impulse from the central government that allowed a resumption of policy developments in Yangzhou. A major institutional change initiated by the central government in 2016 allowed prefecture-level cities like Yangzhou to develop their regulations, something which until then had not been possible. This created the conditions for the adoption of the “Yangzhou Old City Conservation Regulations”, which further legitimized the new policy of urban renewal. Around that same year, the central government also launched a campaign to enhance law enforcement, 4 which forced local governments to become more cautious in the realization of urban projects, as underlined by several interviewees. Even if the OCO has not returned to its central position in recent years, local urban renewal policies continue to develop in the direction of CUR. Signs that this was happening came from one of the districts that governs the Old City, which in 2017 resumed two projects that followed the OCO approach, while experimenting with new approaches to resident participation. The projects were supported by funds from the central government and were initiated by a district political leader. Thus, thanks to the endorsement of the central government and the work of this district-level party cadre, CUR continues to be used in Yangzhou Old City.
Selectiveness, gradualism, or what else?
The radical policy changes introduced in the early 2010s show that it is not possible to speak of a persistent policy style, nor was selectiveness one of the policy style’s major characteristics. Through Howlett and Cashore’s (2009) categories, it is possible to demonstrate that policy change was very radical, affecting the instruments, goals, and other aspects of the urban renewal policy. Following Howlett and Tosun’s theorization (2018), this policy change suggests that a change in policy style occurred, since both instruments and governing ideas changed, also as an effect of the policy transfer. Policymaking actors also changed. With the coming of a new party secretary, thus with the opening of a new political cycle, a new office was given the task to design new policies for urban renewal. Its members included the officials that, a few years earlier, supported the GTZ project, and that thanks to the endorsement of the party secretary could lead the process of reform.
These bureaucrats showed a very proactive policymaking approach, “embedded with bottom-up inputs in local information and knowledge” (Qian, 2018: 203) and drawing on a great deal of foreign knowledge. This was very different from the selectiveness spotted in the earlier phase, which suggested the presence of “domestically dominant institutional practices and preferences” (De Jong, 2013: 99). Even if some of the recommendations offered by GTZ were not adopted during this new political cycle, this non-adoption cannot be ascribed to a cherry-picking approach. On the contrary, this was often explained by the presence of different bureaucratic interests, which forced the OCO to make compromises or even abandon policy projects. This also highlights, once more, that there are no dominant practices and preferences among policymakers, whether they are politicians or bureaucrats. It was not possible to identify proposals incompatible with their preferences, because these preferences were heterogeneous right from the start of the transfer process. Although the incentive structure and promotion rules within the party have remained largely the same, local leaders, both at city or district levels, have shown different visions and preferences towards urban renewal, different attitudes towards policy experimentation, and different preferences in terms of career advancement. And it is precisely their visions and preferences that have dictated the opening or closing of windows of opportunity for the adoption of new policies once they came to power.
The change of party secretary in 2014 led then to a further change in the policy style which, in the second half of the 2010s, became more reactive and responsive to the commands of the central government. As illustrated earlier, the policy developments registered in recent years are related to institutional changes established by the central government. Furthermore, a shift in the policy frameworks of the central government, calling for the conservation of old neighbourhoods, reduced the room for manoeuvre of local leaders when the latter aimed at returning to old urban renewal practices. In the second half of the 2010s, the adoption of policies has also become more sporadic and, to a certain extent, gradual, but once again it is not possible to ascribe this characteristic to the conscious choices of a specific group of policymakers. Without a coordination centre truly responsible for formulating urban renewal policies, as the OCO was in the first half of the 2010s, and without the push of local city leaders, policymaking for the Old City became more fragmented and dependent on the will and capacities of a multiplicity of actors. Therefore, as observed in the latest fieldwork in 2018 and 2019, policy development proceeds slowly through various experiments.
Conclusions
Through a longitudinal analysis of the policy transfer process and its outcomes, and focusing on the “demand side” of policy mobilities (Silvestre and Jajamovich, 2021), this study explored the hypothesis of a distinctive Chinese policy transfer style. For De Jong (2013), this is characterized by selectiveness and gradualism. This study partly invalidated this claim, arguing that there are multiple styles of policy transfer (and more broadly of policymaking) in China. While it is true that selectiveness and gradualism were features of the central government’s policy in the 1980s and 1990s when it introduced the reforms that profoundly transformed China’s economy, governance, and society (Zhou, 2021), the Chinese government, in its Maoist phase as well as close to our times, also demonstrated readiness for bold experiments (Heilmann and Perry, 2011). Furthermore, there is conflicting evidence regarding the results of policy transfers in China – having very limited impacts or, on the contrary, inspiring policy changes (Inkpen and Wang, 2006; Miao, 2018; Zhou, 2021). Therefore, to evaluate policy transfer styles in China, it is necessary to probe its “vast and bureaucratically fragmented political system” (Heilmann and Perry, 2011: 8) in several locations and for protracted periods, and to realize that policy styles can vary considerably, depending on who leads the government, who works in the administration, the resources these individuals have, and the room for manoeuvre these actors possess, in turn, determined by formal and informal structures and various “situational factors” (Freeman, 1985).
This case illustrated that the change in the political cycle linked to the change of local party secretary is central to policy style evolutions. Local party secretaries play a fundamental role in city administrations since they can promote reorganizations, remove, or assign powers to specific departments, and impose political decisions. However, they alone cannot determine the course of policy developments. The longitudinal analysis illustrated that beyond the presence of a proactive bureaucracy, factors external to the local government can also change the power resources of local actors (both policy leaders and bureaucrats). For example, in the most recent phase of the policy development process, although the local party secretary did not prioritize the preservation of the Old City and the work of the OCO, these have been able to continue thanks to the inputs of the central government.
Thus, variations in the combination of factors impacting the policy transfer process, which in turn enhanced or limited the capacities of transfer agents to carry forward the reform process, led to the emergence of different policy transfer styles, at times more reactive or more proactive, at times more selective or, on the contrary, bolder and allowing “for greater bottom-up input than would be predicted from its formal structures” (Heilmann and Perry, 2011: 8). In this complex and highly variable situation, there is one constant aspect aptly identified by Qian (2018: 202): the institutions of the party-state, with the prominence given to party secretaries at various government levels, and the specific role of bureaucrats in policymaking “are critical in understanding policy styles in China”. These, as he illustrates, can change depending on the policy sector concerned. This study demonstrated that a policy style can also change within the same policy sector and that, once again, the combination of party-state institutions and local bureaucrats is fundamental to identifying it.
However, more longitudinal studies of different cases to identify other policymaking patterns are needed. This would not only contribute to understanding the specificities of the policy process in China but also offer elements for comparisons with other contexts sharing similar institutional characteristics. Moreover, studies on policymaking patterns would shed further light on the local dynamics of policy localization (Silvestre and Jajamovich, 2021; Temenos and Baker, 2015), and the factors impacting them. To this end, as also pointed out by Silvestre and Jajamovich (2021), concepts and frameworks derived from public policy studies can help in the analysis of policy mobilities. Beyond the focus on policy styles, this study illustrated how the use of Howlett and Cashore’s (2009) categories of policy change helped reveal “how policies are ‘arrived at’ in particular places” (Temenos and Baker, 2015: 842), how best practices are concretely translated into policies, and what degree of policy change they inspire. Further borrowings from policy studies are therefore encouraged to enrich our knowledge of policy mobilities.
Footnotes
Acknowledgements
The author would like to thank Claire Colomb, Guillermo Jajamovich, Sergio Montero, Farad Mukhtarov, Ryan Whitney and the “Policy Mobilities Working Group” for the comments provided in earlier versions of this paper, as well as the editor and the two anonymous reviewers for their constructive suggestions and comments. The responsibility for the contents of the document lies, however, entirely with the author. The author would also like to thank the French Ministry of Europe and Foreign Affairs and the French Ministry of Higher Education, Research and Innovation for having supported this research via the exchange programme “Cai Yuanpei”. Thanks also to the “IN-EAST School of Advanced Studies”, an initiative funded by the German Federal Ministry of Education and Research, for their support in the research period 2018–2019.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
