Rethinking China’s urban governance: The role of the state in neighbourhoods, cities and regions

Following the notion of the entrepreneurial city, this paper examines recent scholarship about China’s urban governance. Despite prevailing marketisation, the role of the state is visible in neighbourhood, cities and city-regions. The state necessarily deals with a fast changing society and deploys market-like instruments to achieve its development objectives. Through multi-scalar governance, the state involves social and market actors but at the same time maintains strategic intervention capacity. China’s contextualised scholarship provides a more nuanced understanding beyond the entrepreneurial city thesis, which is more state-centred.


I Introduction
China's phenomenal urbanisation is of worldhistorical significance and imposes profound theoretical and policy challenges. It first challenges our understanding of the nature of cities, contemporary urbanisation and urban transformations, as well as the model of urban governance in a wider geographical perspective (Robinson, 2006). Against the worldwide transition towards neoliberalism (Harvey, 2005), urban China provides an excellent and timely case to examine its specific development approach in a wider theoretical debate of entrepreneurial governance. This paper will rethink China's model of urban governance through cross-examining empirical studies and will pay attention to the particularities of Chinese neighbourhoods, cities and city-regions.
The research gap is the lack of in-depth and concrete understandings of the role of the state on the ground in diverse neighbourhoods, cities and regions. Very few studies on Chinese urbanisation simultaneously address urban governance across these levels. This paper reviews a growing body of literature on China's governance under the initial hypothesis of state entrepreneurialism which combines planning centrality, market instruments and observable social agencies (Wu, 2018b). We will rethink China's governance model through systematic and multi-scalar investigation. This study will expand our dichotomous understanding of the Chinese city beyond market-driven agglomeration or state-led urbanisation. The aim is not to propose a distinct Chinese model but rather to provide a nuanced understanding beyond either the authoritarian state or neoliberal urbanism.
Against an established literature on the 'new urban politics' which has appeared conspicuously under urban entrepreneurialism (MacLeod, 2011), this paper critically rethinks various models of urban governance derived from Western contexts. These models capture some salient features of contemporary urban transformations. However, they need to be enriched and complemented with mid-level conjunctional analysis which pays attention to both the geographical particularities and the general political economic processes of capitalism (Peck, 2015). The original notion of governance as 'governing without government' through self-governing networks (Rhodes, 1996), urban development through policies centred upon urban development corporations (Imrie and Thomas, 1999), and managing mega-urban projects through market contracts (Raco, 2014) need to be carefully examined in the Chinese context. Jessop's (2016) state-theoretical perspective highlights the critical role of the state in securing governance coherence.
The central concern is the role of the state. To be clear, the state here refers to the Chinese party-state. The distinctive feature is the close association between the government and the Chinese Communist Party (CCP) ( (Saich, 2015); (Thornton 2013)). The party maintains the vertical control and cadre appointment (Cartier, 2015;Chien and Woodworth, 2018). The central government maintains the 'territorial strategies' through the political system of the Chinese party-state with administrative rank of officials within the jurisdictional hierarchy (Cartier, 2015). Thornton (2013) reveals that the party has recently advanced into urban grassroot society and developed a relationship with various non-government associations and organisations. In both urban villages (Kan and Ku, 2021) and urban business districts (Han, 2015), the party has deepened its territorial reach and strengthened party building. In this paper, we address the role of party-state in urban governance without specifically interrogating the internal dynamic between the state and party because of the close association between these two entities. In other words, the notion of the state here does not exclusively refer to the government but rather a hierarchical and multi-scalar party-state.
We ask whether the turn to market coordination has constrained the conventional political processes as shown in Western democratic societies. For example, the 'post-political city' describes a new politics of exclusion and substantial democratic deficit (Swyngedouw, 2009), which implies a declining role of the state. Combined with growth, entrepreneurialism and privatism, the post-political urban model leads to 'authoritarian' and undemocratic characteristics. Such a trend shows, instead of state retreat, state-led privatisation (Raco, 2014;Whiteside, 2021). At the neighbourhood level, the study of micro-politics in 'master-planned residential estates' reveals rising private governance (McGuirk and Dowling, 2009). At city-regional levels, 'rescaling' reflects the state's spatial selectivity towards 'territorially extensive configurations' (Brenner, 1999(Brenner, , 2004. More recently, the urban governance model has evolved from urban entrepreneurialism to financialising urban governance (Peck, 2017). The play of growth machine politics is now transformed into a new generation of debt-machine dynamics (Peck and Whiteside, 2016). Urban governance is imposed by financial imperatives, in particular the financialisation of the home (Aalbers, 2008), leading to the financialised city (Halbert and Attuyer, 2016). The notion of financialised governance seems to suggest a weaker role of the state in comparison with financial actors.
Further, the study of urban governance has expanded from the earlier focus on governing urban development under entrepreneurialism to wider studies of the governance of urbanism and everyday social life ranging through citizenship, political representation, identities and living spaces (McCann, 2017). Post-structural studies on governance indicate that the models of governance may not be simply determined by economic conditions but may also be developed through policy mobility and the travel of 'best practices' (McCann and Ward, 2011), as well as inter-references between different places (Roy and Ong, 2011). Individually, these models provide inspiration to Chinese research on urban governance (e.g. the mobility and mutation of entrepreneurialism to small Chinese cities, He et al., 2018; 'policy sprawl' of suburban new towns, Miao and Phelps, 2021). But there has been no systematic attempt to provide an overall picture of China's model of urban governance (Wu, 2018b), despite some critical reflections on the applicability of neoliberal urbanism (Buckingham, 2017;Lim, 2014;Peck and Zhang, 2013;Robinson et al., 2021;Wu, 2010;Zhou et al., 2019). The examination of Chinese urban governance here thus needs to go beyond economic development and interrogate the governance of everyday living spaces in neighbourhoods, urban development strategies and environmental discourses in cities, together with city-region planning and inter-city collaborations.
The paper rethinks China's urban governance through its transcendence beyond the literature of entrepreneurial governance and interrogates empirical works by China scholarship on these three levels neighbourhoods, cities and regions. Each level sees the development of multi-scalar governance led by the state. In other words, we regard them as a new state space in which actors across different scales are interrelated and interact (Shen et al., 2020). Connecting these streams of literature, we focus on the role of the state. In neighbourhood governance, the paper unravels the role of the state in the everyday living space that interfaces with the society. In urban governance, we reveal that the local state strives to balance economic interests in the land market and development strategies under 'national political mandates' ). In regional governance, we uncover the state's role in region-building, managing both the expansion of the central city and economic agglomeration in city-regions. The paper thus contributes to the theorisation of urban and regional governance.
This research on China's urban governance has significant policy implications. The review of Chinese urban governance is timely for the implementation of an UN-endorsed new urban agenda in China and for China to meet the challenges brought by global trade tensions. The paper helps to recommend how China should change its development model and how the outside world should help China to address its immense challenges from urbanisation.
In short, this paper aims to rethink China's model of urban governance beyond the characterisation of neoliberal urbanism or state authoritarianism and to provide a nuanced understanding of the role of the state in the dynamic relation of state-market-society (He and Lin, 2015;He and Qian, 2017;Wu, 2018bWu, , 2020aYe, 2018;Yeh et al., 2015b) in its governance through neighbourhood, urban and regional levels and multi-site examinations, and to assess the implications for urban theory and policies. The paper is structured as the following five sections: a review of China's development approaches, followed by examinations of neighbourhood governance, urban governance and regional governance, and finally theoretical implications and policy agenda.

II China's development approaches
In this section, we review China's approaches to economic and urban development and the debates over the role of the state. In particular, we highlight the response of the state to a fast changing society and decentralised economic decision-making. China has experienced sustained economic growth since the economic reform of 1979. Its growth accelerated after joining the World Trade Organisation (WTO) in 2001. The Chinese economy seemed resilient during the Global Financial Crisis in 2008. Now, confronted with intensified trade conflicts and international politics in the post-pandemic era, it is time to rethink China's development approaches and its governance model. In fact, there have been debates about whether China has adopted a distinctive governance approach to economic development, a Chinese governance model. The market-oriented reform started in 1979. But Deng Xiaoping's southern China tour in 1992 led to a more fully fledged export-oriented development which has become more established since China joined the WTO in 2001. China has become the 'world factory' and engages with and is dependent upon the global economy. Harvey (2005) described China's model as 'neoliberalism with Chinese characteristics' or specifically 'neoliberal authoritarianism'. In urban redevelopment, a 'neoliberal' approach -'property-led redevelopment' (He and Wu, 2005) was adopted to encourage real estate-driven old neighbourhood regeneration (He and Wu, 2009), leading to gentrification and mixed uses. Overall, marketisation was promoted in housing, land, labour and social provision (Walker and Buck 2007).
Recent studies have shown that the state operates through land-based finance to develop the 'worldfactory'. The entrepreneurial local state is a key actor in this business model. Incentivised by a tax-sharing fiscal system, the entrepreneurial local state uses its monopolistic control over land to attract foreign investment, which in turn utilises the migrant labour force. There are now extensive studies on China's land-based finance model (Lin, 2014;Su and Tao, 2017;Tao et al., 2010;Wu, 2019). Su and Tao (2017) tried to explain the 'institutional roots' of China's developmentalism. Because of local fiscal shortages, local governments tried to maximise local revenue through leveraging infrastructure and real estate development for industrial expansion. Buckingham (2017) stresses that this model is built upon the legacy of socialist institutions which divides urban and rural areas. This governance model continues to produce the social exclusion of migrants (Buckingham, 2017;Solinger, 1999).
The visible role of the state triggers debate over whether neoliberalism is applicable to China. Peck and Zhang (2013) stress the China model as a variegated capitalism, while Wu (2010) believes China in practice uses market mechanisms without adhering to a neoliberal ideology. Zhang and Peck (2016) further suggest that Chinese capitalism consists of a variety of sub-models based on different regions. There is no single governing model in the geographical sense. Historically, the Chinese central government has shifted governing strategies at major junctures since the economic reform (Lim, 2014(Lim, , 2019. Still, there is a long-term research interest in generalising the overall Chinese governance approach. Keith et al. (2013) challenge the neoliberal interpretation of China's model and argue that China is not evolving towards an individualised neoliberal economic life due to its cultural and social traditions. Buckingham (2017) argues that Mao's institutions such as the hukou system were revived to manage the migrant population and create a labour force for economic development. Overall, the neoliberal city thesis is debated in Urban Studies and found to be contradictory in various contexts (Le Galès, 2016;Parnell and Robinson, 2012;Pinson and Journel, 2017). In China, the use of neoliberalism to explain China's development approaches is debatable (Buckingham, 2017;Wu, 2010;Zhou et al., 2019). The visible role of the state and lingering state institutions contradict the expectation of a declining state. On the other hand, active market development and participation in global economies is undeniable. To reconcile these seemingly contradictory tendencies, the notion of 'state entrepreneurialism' is proposed to characterise the simultaneous application of 'market instruments' and 'planning centrality' (Wu, 2018b). Further, Zhou et al. (2019) argue that current studies on neoliberalism do not provide an adequate description of China's governance and neglect 'innovative practices' in China, for example, a 'spontaneous reconfiguration of the urbanising society' at the grassroots due to new technology and a state that is interventionist in innovation. Similarly, Zhang and Wu (2019) highlight the state's role in high-tech park development while actively associating with global production networks.
While the visible role of the Chinese state is widely noted, the Chinese state must face a changing society and more decentralised and local economic decisions in the post-reform era (Logan, 2018;Mertha, 2009), which increasingly presents a multiscalar nature (Lim, 2019;Shen et al., 2020;Wu, 2016).
First, a growing body of literature reveals social agencies beyond state and market relations (He and Lin, 2015;He and Qian, 2017;Liu and Yau, 2020;Liu et al., 2015;Smart, 2018). Consumption and consumerism are notably new forces for economic growth and governance changes (He and Lin, 2015;Liu and Yau, 2020;Qian and Lu, 2019). For example, Smart (2018) highlights the importance of interpersonal interactions and cultural interpretations of Chinese urban development. The discrepancy between formal governance structure and 'informal responses both in society and within government' reflects an important feature of social agencies in China's model of governance (Smart, 2018). Liu et al. (2015) reveal that migrants in urban villages form trans-local social and business networks and develop their space for business and living in the city. The active role of social agencies refines our understanding of 'authoritarianism'. Post-reform China has seen negotiation between the state and market and the development of social organisations (Saich, 2000;Mertha, 2009), the development of nongovernment organisations and the re-established link with the party (Thornton, 2013). To reflect social actors, Mertha (2009) added non-public actors to the framework of 'fragmented authoritarianism' to explain the input of society in governance. Similarly, Lee and Zhang (2013) proposed the model of 'bargained authoritarianism' to describe the foundation of governance based on market exchanges, rule-bound games and social bonds. The concern over social stability raises the influence of society. On the other hand, these bargaining mechanisms allow the state to absorb popular protests at the grassroots.
Second, related to growing social actors and diversity, Qian and An (2021) stress the importance of everyday urbanism in China's development and governance. People's responses to macrostructural changes as well as their desires for lifestyles and identities are productive forces for urban governance. For example, the desire for homeownership adds to the force of housing financialisation as a process of assetisation  in which the state recognises this dynamic and uses housing commodification and homeownership to create the impetus for economic development. Rising consumerism and greater desire for residential privacy and better environmental quality represent changing social trends (Zhang, 2010) with which the state has to deal in its governance.
Third, rather than considering the state as a single entity, recent studies on China's urban governance highlight the state's multi-scalar nature and complex central and local interactions (Li and Jonas, 2019;Lim, 2019;Shen et al., 2020;Wu, 2016). The state is territorially organised but also transcends geographical spaces (Cochrane, 2007;Jonas, 2020;Robinson et al., 2021). There are complex central and local relations and different hierarchies of government and inter-local competition (Cartier, 2015;Logan, 2018;Lim, 2019;Zhang and Wu, 2006). Thus, it is more appropriate to think of the state apparatus as an assemblage (Robinson et al., 2021;Shen et al., 2020). There are negotiations across scales. For example, large mega-urban projects are often driven and delivered by multiple state agencies (Shen et al., 2020;Wang and Wu, 2019;Yang et al., 2021).
Instead of applying an existing governance model to China to suggest that the Chinese version is a variegated version of the entrepreneurial city seen in the West, we argue that it is important to review alternative interpretations of the Chinese urban governance model, provide a more nuanced understanding, and detail its operation through multi-scalar investigation. In the following sections, this paper examines the role of the state in neighbourhood, cities and regions.

III Neighbourhood governance
In Western market economies, the emergent 'privately governed neighbourhood' is a salient feature, which raises great concerns over the decline of the public realm (Low, 2003;McGuirk and Dowling, 2009). Typically, these are 'gated communities' (Blakely and Snyder, 1997;Le Goix and Webster, 2008) or 'master-planned neighbourhoods' (McGuirk and Dowling, 2011), for a variety of reasons: lifestyle choices and common interests, residential segregation and a deteriorating feeling of security (e.g. 'ecology of fear', Davis, 1990;or dystopia, Low, 2003). More broadly speaking, from the governance point of view emerging gated communities represent 'urban secession' through which wealthier households relocate to the suburbs and disassociate themselves from the central city (Keil, 2000), increasing private governance (McKenzie, 1994(McKenzie, , 2005 through which residents 'vote with their feet' into neighbourhoods that fit their consumption needs. Gated neighbourhood governance is also regarded as a 'club realm' (Webster, 2002) where services are provided by the market according to the consumption preferences of a group of residents. Further, private governance represents 'suburban neoliberalism' (Peck, 2011), in which the state retreats from service provision and market provision prevails. However, as shown in master-planned residential estates in Australia, there is a wide spectrum of the politics of social reproduction and the self-regulating consumer citizen (McGuirk and Dowling, 2011).
Market-oriented economic reform in China has generated profound impacts on neighbourhood governance. So-called 'communist neo-traditionalism' (Walder, 1986), built upon work-units' collective consumption, has declined. The end of state work-unit housing provision and the introduction of housing markets have shifted responsibility for service provision from state work-units to local governments and finally to property market companies (Wu, 2002(Wu, , 2018a. The residential neighbourhood has become a basic unit for organising urban life (Wu, 2002).
The introduction of housing and land markets has made it possible for urban residents to buy housing in 'enclosed estates' (Pow, 2009;Wu, 2005), forming 'enclave urbanism' (Wissink, 2019). With greater residential privacy and autonomy in housing consumption (Huang, 2006;Lu et al., 2020;Pow, 2009;Wu, 2005;Zhang, 2010), Chinese neighbourhoods have seen a new space participated by the society in terms of neighbourhood governance. China has seen a diversity of neighbourhood types, ranging from more formal work-unit housing compounds to exmunicipal housing estates, from newly built 'commodity housing' estates, to informal settlements like urban villages or private housing neighbourhoods developed before 1949. Because of the diversity of neighbourhood types in China, there is no universal model of neighbourhood governance (Wang and Clarke, 2021). In commodity housing estates, homeowners' associations are developed (Fu and Lin, 2014;He, 2015), while in migrant-concentrated enclaves, either villagers' committees or shareholder cooperatives play an important role (Po, 2012;Xue and Wu, 2015). Neighbourhood governance changes across different residential types (Lu et al., 2020;Phelps et al., 2021;Wang and Clarke, 2021). Lu et al. (2020) reveal the contrast between relocated and resettlement housing and upper-market housing neighbourhoods. Wang and Clarke (2021) describe four models: collective consumption, service privatisation, civic provision and state-sponsored governance, and they reveal diversity and complexity in neighbourhood governance due to the path-dependent features of individual types of neighbourhoods. While some neighbourhoods have seen service privatisation and hence greater collective decision-making by residents, for old urban neighbourhoods local state organisations have to sponsor or directly organise their social services (Wang and Clarke, 2021;Wu, 2018a). Phelps et al. (2021) similarly contrast different everyday lives and work relations in the foreign industrial dormitory and state work-unit neighbourhood. The scholarship on neighbourhood governance in China demonstrates different mechanisms in diverse neighbourhoods: for example, migrant-concentrated neighbourhoods versus middle-class homeowners' neighbourhoods, which have different implications for migrants and homeowners. Overall, we have seen activism and some self-governance mechanism through the homeowners' association in formal housing neighbourhoods (Cai and He, 2021;Fu and Lin, 2014;Wang and Clarke, 2021) and weaker selfgovernance in informal migrant enclaves where migrants are largely excluded from the process (Wu, 2012). Now, gated residential areas are managed differently from the management of work-unit housing estates (Pow, 2009;Zhang, 2010;Tang, 2018). Property management companies take responsibility for service delivery. In middle-class housing estates, homeowners' associations (HoAs) are set up, becoming a new agency of governance (Fu and Lin, 2014;He, 2015). There are signs of homeowners' activism (Boland and Zhu, 2012;Cai and Sheng, 2013;Shin, 2013). The homeowners' association provides a counterbalance to market forces and state regulatory interference (Fu and Lin, 2014). To deal with diversifying and rising social interests, the state tried to rejuvenate the residents' committee through its 'shequ campaign' (Read, 2012;Shieh and Friedmann, 2008;Tomba, 2014) and more recently the establishment of 'grid governance' (Tang, 2020). The latter divides residential neighbourhoods into geographical and administrative grids. As a 'social management innovation', these grids involve multiple social actors and government officials assigned with responsibility duties.
Nevertheless, the development of gated estates may not lead to private governance . New governance structures are reinvented and strengthened by the state to maintain its governance over grassroots organisations (Cai and He, 2021;Tang, 2020;Wu, 2018a). Huang (2006) stresses a continuation of 'collectivism' and collective control in these gated places. To the state, residential neighbourhoods are important because they serve as a space where governance is 'socialised'that is, face-to-face politics is set up as a mechanism of control and governance (Woodman, 2016). Further, Wissink (2019: 183) argues that Chinese enclave urbanism does not represent self-governance space but rather comprises 'assemblages of heterogeneous elements that are themselves part of multiple assemblages operating on various "scales" and thus have multiple determinants'. There are broad linkages, social interactions and social networks between enclaves and the outside world.
Despite active social agency, everyday urbanism has not evolved into a strongly self-governed society which can be properly labelled civil society. The state has strengthened the infrastructure of governance infrastructure (Cai and He, 2021;Tang, 2020;Wu, 2018a). Recent research shows that it is imperative to examine not only entrepreneurial economic strategies but also the state's deliberate actions to maintain governance coherence, for example, in the governance of informal space and street vending (Huang et al., 2014), or in the tension between the state and society (Shin, 2013). For migrants, preliminary evidence shows that in urban villages, through developing extensive social networks, migrants adapt within the enclave, formulate a flexible but effective production system and consequently become migrant entrepreneurs (Liu et al., 2015). However, the participation of migrants in social affairs in villages and their roles in governance are still constrained. As for gated communities, Chinese gated communities may not resemble rising private governance (Webster, 2002). Homeownership leads to embryonic participatory governance and social innovation. With varying residential diversity in different neighbourhoods, self-governance varies in neighbourhood governance (Wang and Clarke, 2021) and has specific and limited effects .
Chinese neighbourhoods demonstrate active social agencies. Millions of migrants and urban residents adapt to their constraints and opportunities and transform urban China through their agency (Liu et al., 2015). In the redevelopment of urban villages, the bargaining power of villagers forced the state to make compromises and provide more compensation to villager collectives (Lin, 2015). Shin (2013) documented extensive property rights activism in middle-class neighbourhoods. In the gentrification of Shanghai's historical neighbourhoods, Arkaraprasertkul (2018) discovered that the original residents were themselves active actors as they discovered the economic value of heritage and renovated their vernacular housing for rental, which led to the diversification of the neighbourhood. Such bottom-up neighbourhood change is driven through intervention by local government or real estate developers. Similarly, in the redevelopment of urban villages in Southern China, local villager collectives or shareholder cooperatives play a very important role. In the development of peri-urban new towns such as Guangzhou University Town, university students are the new consumers and residents of the place (Liu and Yau 2020). They further attract speculative affluent investors, leading to the gentrification of former rural places. Lin and Kao (2020) reveal the role of grassroots environmental activism to counter state projects.
Facing an increasing role of society and diverse social actors, the role of the state continues to be critical as it strives to manage the migrant population (Buckingham, 2017) and initiate urban redevelopment . We illustrate these roles through recent strategies to 'incorporate migrants' and to promote urban redevelopment rather than wholesale demolition. However, as shown below, these strategies, often initiated separately, can often be contradictory. First, the governance of migrants represents the dual motivations of control versus developing the domestic market. The social agency of migrants is not entirely prohibited, demonstrating a perplexing politics of inclusion and exclusion. Recently, in response to managing migrant population and environmental pressure, China's new urbanisation plan (2014-2020) requires a change from land-centred to people-centred urbanisation beyond GDP-ism. Just as with immigrants in Europe (Wacquant, 2008), migrant social incorporation is considered the single most significant social challenge for future China. While promoting migrant social incorporation, the state also strives to govern migrant urbanism. Recently, Beijing and Shanghai as well as many other Chinese cities initiated large-scale urban renewal and demolition of informal spaces, which displaced migrant populations (Wong et al., 2018;Wu et al., 2013). These new urban strategies cannot be exclusively represented as the entrepreneurial city . For example, for Beijing, governing migrant urbanism is associated with the role of Beijing as the capital and the state's consideration to eliminate congestion and improve air quality. Hence regeneration involves dispersing, regulating and upgrading (Wong et al., 2018). Nevertheless, in historical neighbourhoods, 'incremental regeneration' (weigaizao) under the mandate of heritage preservation is promoted .
Second, the politics of urban redevelopment does not follow the path of the growth machine in North America. The case of Guangzhou shows that, although the speedy redevelopment of urban villages was initially intended for hosting the Asian Games in the late 2000s, soon after the completion of its pilot project (Liede village), the state began to limit the density and intensity of development. In other words, the growth machine was abolished (Guo et al., 2018). Political considerations then prevailed. The underlying motivation of the state is not to pursue land profits but rather a 'strategic' consideration of the 'interests' of the whole city. Similarly, in Shanghai's waterfront redevelopment, the policy discourse is that 'it is for the people' (Li and Zhong, 2021). In Nanjing, heritage preservation created significant impacts on old neighbourhoods. But this more historically sensitive regeneration is often the result of social contestation and elite participation  under the shifting ethos of the state to avoid real estate speculation and maintain social cohesion. Recently, the central government encouraged incremental regeneration and community participation in dilapidated neighbourhoods. New redevelopment projects are strongly influenced by state politics and political mandates . Although these projects are associated with the property market, they cannot be exclusively or appropriately explained as an outcome of growth machine politics. In the long term, it remains to be seen whether the new trend of neighbourhood governance with greater social and market agencies may eventually transform state-centred governance in China. Harvey's (1989) seminal thesis of 'urban entrepreneurialism' explains the transformation of urban governance from managerialism to entrepreneurialism in late capitalism. The thesis has been analytically developed into the 'entrepreneurial city' (Jessop and Sum, 2000) and neoliberal planning (Sager, 2011). However, Jessop (2006) argues for a more state-theoretical perspective, as these entrepreneurial endeavours are highlighted by a series of state actions, discourse and self-images. Rather than thinking laissez-faire entrepreneurialism, Jessop and Sum (2000) actually pointed out a series of stateorchestrated place actions, diplomacy and strategies. In other words, the entrepreneurial city is no longer driven only by the growth machine based on real estate interests but also represents municipal statecraft, which includes multiple governance agendas, experimental public interventions and inter-urban diplomacy (Lauermann, 2018), active scalar fixes through state rescaling (Bok, 2019), varieties of the new urban managerialism (Phelps and Miao, 2020), and financialised statecraft (Pike et al., 2019) or state entrepreneurialism using market toolkits (Wu, 2018b(Wu, , 2020b(Wu, , 2021.

IV Urban governance
The concept of the entrepreneurial city has been applied to Chinese cities, including large cities such as Shanghai (Wu, 2003) as well as remoter, smaller cities striving to attract overseas investment (Chien, 2013;Chien and Gordon, 2008;He et al., 2018;Su, 2015). Chinese local governments demonstrate their willingness to engage in market-like entrepreneurial activities. In the study of Chinese local development, the career motivation of local politicians and the central-local fiscal relation are two main explanations (Guo, 2020). The thesis of the 'GDP tournament' emphasises how local political leaders compete over targets for GDP growth set by the central government in order to gain career promotion (Li and Zhou, 2005). GDP-ism thus is a major driver for entrepreneurial governance, especially at the level of the local state. On the other hand, a more structural explanation stresses fiscal structure and local revenue balance (Hsing, 2010;Su and Tao, 2017;Tao et al., 2010). Hardened fiscal responsibility accompanied with decisionmaking autonomy turned local governments into organisations similar to 'industrial firms' (Walder, 1995). The tax-sharing system requires local government to seek its own revenue sources, and statecontrolled land institutions allow the local government to operate land development to generate profits from land sales to fill the gap in local expenditure. Tao et al. (2010) explained a quite sophisticated dynamic of land value capture: through promoting industrial development to generate overall demand for housing and commercial land but at the same time controlling land supply, local governments push up the price of residential and commercial land sales to capture land values. This mechanism incentivises the local government or, more precisely, local state-owned corporations (so-called investment and development platforms) (Feng et al., 2021), to invest in further infrastructure development. The operation in essence requires urban development, which has further evolved into the form of new cities and towns which combine both residential and industrial functions (Hsing, 2010).
The mechanism of 'land-based financing' (Lin, 2014;Tsui, 2011) is arguably the most important driving force for local governments to aggressively promote urban development (Wu, 2015). Following the thesis of the growth machine initially developed in the United States (Logan and Molotch, 1987), an emerging literature tends to explain Chinese urban development through urban entrepreneurialism (Lin, 2014;Wu, 2015). But Phelps and Miao (2020) argue that not all growth incentives come from land. The thesis of the local state's 'GDP tournament' is actually a political explanation, focussing on the behaviour of local officials and their interaction with upper governments. As Guo (2020: 225) points out, building performance and seeking promotion are the internal impulses for Chinese local state leaders to pursue entrepreneurialism; on different occasions when the central government's strategy has shifted from growth first to more 'people-oriented' growth (Li and Zhong, 2021) and ecological civilisation , the evaluation of performance may well be different. The entrepreneurial city thus may not be simply oriented towards economic growth or revenue maximisation. In this context, the local state may perform the ecological fix rather than pursuing growth at any cost .
Land-driven development and revenue maximisation are seemingly plausible explanations for the cause of the entrepreneurial behaviour of the local state at specific historical moments. However, the model leaves out the 'politics of development' or geopolitics (Jonas, 2020a(Jonas, , 2020b and intentional strategies of the state, using the market only as a tool rather than replacing its rationality of governance (Wu, 2020b). It is therefore important to study the governance of actual development processes in which revenue maximisation might be only part of the motivation (Wu, 2018b), and also the 'endogenous impulses of the state' (Guo, 2020). The local government may not act 'like a firm' (Walder, 1995), depending upon complex considerations to pursue territorial interests versus aligning with central government policies. In short, considerations other than economic growth, for example, to maintain 'planning centrality', may be imperative in urban governance (Wu, 2018b). To what extent does Chinese emergent urban politics resemble the 'new urban politics' in the literature (Cox, 2016)? Existing studies reveal many distinctive characteristics and specific approaches, implying a Chinese model of governance. Though informative, current studies on Chinese urban governance seem to reveal a great variety (He et al., 2018;Zhang and Peck, 2016) as well as deviations and failure (Xue and Wu, 2015). Current research on China's urban governance is often focused on specific themes such as land development and financialisation (Lin, 2014;Tao et al., 2010;Wu, 2019), which often implies that the state has been captured by capital (Cochrane, 2007). However, it is imperative to examine the nature of Chinese urban governance with an explicit statetheoretical view (Jessop et al., 1999;Jessop, 2016). Moreover, it is useful to examine diverse cities (large and small, in coastal and inner regions, state projects and traditional small towns) because there might not be a singular Chinese 'model' but quite a range of historically and geographically contingent practices (Zhang and Peck, 2016).
Recent studies on China's urban governance have made progress in understanding concrete mechanisms and actors such as urban development corporations (Feng et al., 2022;Jiang and Waley, 2020;Li and Chiu, 2018;Wu, 2021). These studies demonstrate the close relation between market actors and the state, as urban development corporations are actually state-owned enterprises (SOEs) owned by various governments. As a result, the state manages to steer development through its development actors.
The role of the state in urban governance is not limited to a passive status determined by capital accumulation. In urban redevelopment, Li and Zhong (2021) show that waterfront redevelopment in Shanghai is more state-centred. Powerful state actors considered urban regeneration as a way of enhancing residents' access to the waterfront. But this development objective has been achieved without much public participation. In this way, the local state acts quite differently from a growth machine. As mentioned earlier, China's recent urban regeneration campaigns are beyond the local politics of the growth machine and are heavily guided by state politics . The local state uses regeneration projects to achieve extra-economic objectives, although the actual operation may utilise market actors and tools.
The strong role of the state can also be seen in environmental governance. Confronted with environmental pressures and climate change, the Chinese central government promotes 'ecological civilisation' and ecological urbanism. The planning of ecocities is criticised for eco-branding because these developments lack sufficient environmental and social considerations (Caprotti et al., 2015;Chien, 2013;Pow and Neo, 2013;Wu, 2015). The new practices, however, are not necessarily determined by entrepreneurial governance but rather in the context of a more regulatory state, for example, with stronger control over informal space, ecological quality and carbon emissions. This rising state environmentalism under Chinese urban governance has been seen as a shift from the 'sustainability fix' deployed by the entrepreneurial city (While et al., 2004) to an 'ecostate' (While et al., 2010), as shown in a more ecologically oriented enforcement . Entrepreneurial governance is transformed to cope with the problems derived from earlier marketisation and entrepreneurialism such as environmental degradation. In environmental governance, the role of the state deviates from that of a growth promoter and has to balance local economic interests and environmental targets which are set by the central government.
In short, although entrepreneurial governance is a useful way to characterise the economic consideration, the concept is too general but also too restrictive, lacking nuance and specificity in the Chinese context. It does not give sufficient attention to the Chinese state in urban governance. The state may not be captured by (external) capitalist interests because the dominant class is not a capitalist class. The state represents a 'revolutionary' class, and its legitimacy is not defined by capitalist political processes (Shue, 2018), which means that the intention to introduce governance may not be exclusively economic and profit-oriented but rather have broader aims including greater social stability (Lee and Zhang 2013). Second, entrepreneurial governance stresses market actors. However, the role and strength of the society needs to be recognised (He and Qian, 2017;Logan, 2018;Smart, 2018). Once marketisation started and now it has come to prevail, a vibrant society outside the capitalist interest may be formed and resist or reverse the logic of capital accumulation (Qian and An, 2021). Migrant agencies in urban villages demonstrate that space is not under the total control of either the state or capitalist logic. As the state is not able to control every actor in the regime of urban development, it must deal with diverse interests through interaction with the market and society . Hence, it is in this sense that a governance approach has been formed beyond the government (Rhodes, 1996).
The study of Chinese urban governance has led to quite fruitful comparisons between manifestations specific to China and theoretical explanations derived from the West. For example, there are interesting comparisons between entrepreneurial governance and neoliberalism (Harvey, 1989;Peck, 2011), a variety of new public management and municipal state agencies (Lauermann, 2018;Phelps and Miao, 2020), new municipalism and social contests (Janoschka and Mota, 2021;Whiteside, 2021) and 'state entrepreneurialism' (Wu, 2018b(Wu, , 2020b. This paper reveals not only the institutional reasons for local entrepreneurial governance, which are specific to China, but also the possibility that Chinese urban governance may go beyond urban entrepreneurialism due to various political motivations and politics of development (Wu, 2020b), especially state strategic considerations including social stability (Lee and Zhang 2013), environmental crises  and financial risks (Feng et al., 2021;Wu, 2021;Wu et al., 2020).

V Regional governance
Under the globalisation of economies, advanced market economies in the West have witnessed a new spatial form known as the 'global city-region', 'mega-urban regions', or 'megaregions' (Hall and Pain, 2006;Harrison and Hoyler, 2015;Scott, 2001Scott, , 2019. There are different interpretations of the dynamics and associated governance models. For some, this can be seen as 'post-suburbia' (Phelps and Wood, 2011;Teaford, 1997) in which a new suburban economy is extended into the city-region in the aftermath of mass residential suburbanisation (Keil, 2018). This is arguably associated with the perceived demise of the nation state and market prevalence in governance. Decentralisation and polycentric urban forms are seen as the spatial manifestations of 'suburban neoliberalism' (Peck, 2011). Brenner and Schmid (2014) conceptualise metropolitan growth and beyond as 'planetary urbanisation', while Keil (2018) attributes it to an extended form of urbanisation after suburbanism, paying attention to a wide range of peri-urban and exurban development in the global South. There are debates over the key driving force for a new governance model of city-regionalism. From the perspective of agglomeration, economic geographers focus on inter-firm linkages and related production networks (Scott, 2001;Scott and Storper, 2015;Storper and Scott, 2016). Others, however, argue for a perspective of geopolitical process for city-regionalism (Jonas and Pincetl, 2006;Jonas et al., 2014;Jonas and Moisio, 2018;Li and Jonas, 2019). A major difference between these two paradigms is that the latter stresses geopolitics, in particular politics dealing with the social reproduction of capital accumulation. Jonas and Pincetl (2006) point out that the new regionalism is created through the 'politics of distribution' and argue that a 'new civic regionalism' in the U.S. has been rolled out by those who have regional business interests, which has led to the reorganisation of local and state power in cityregions. Rebutting city-regionalism as an outcome of globalisation and agglomeration, Jonas et al. (2014) argue that the formation of city-region governance is a process of scale building. Harrison and Hoyler (2015) suggest that megaregion building involves both state and non-state actors. Focussing more explicitly on scale politics, Brenner (2004) conceptualises 'state spatiality' at the city-region level. The emergent city-region is due to the selection of this particular city-region scale for governing economic development and managing social cohesion. On the other hand, Roy (2009) argues that the concept of 'global city-regions' is rooted in the Western experience and is thus unable to cover the multiple forms of metropolitan development in the global South.
China offers an excellent and exciting case for understanding regional governance (Jonas, 2020a;Li and Wu, 2012;Wu, 2016;Ye, 2018;Yeh and Chen, 2020;. The development of regional governance demonstrates two processes: economic regionalisation through the development of interconnected regional economies and state rescaling. The former is regionalisation through associated economies and regional infrastructure . The latter refers to territorialised governance at a regional scale, or regionalism. Thinking about governance at the regional scale, that is, regionalism, we need to understand the change in governance in response to regionalisation. China's spatial pattern of urbanisation has been highly uneven, with concentration in the coastal region and especially the Pearl River Delta (PRD), the Yangtze River Delta (YRD) and the region of Beijing-Tianjin-Hebei (Jing-Jin-Ji). Overall, regional governance presents three spatial levels: the first is the jurisdictional area of the municipality centred on the core central city and towns under its jurisdiction (like the Shanghai metropolitan area). The second is integrated cities comprising two adjacent jurisdictional areas (like Guangzhou-Foshan). The third is mega-city regions with multiple metropolitan areas (like Jing-Jin-Ji, YRD and PRD). The emergent term 'urban clusters' (chengshiqun) loosely refers to the mega-city regions or is sometimes translated in more general terms as city-regions.
Globalisation and foreign direct investment have led to 'ex-urbanisation' in the PRD (Sit and Yang, 1997). The development of high-speed trains, intercity railways and cross-border metro lines have significantly reduced travel time and improved the connectivity of a vast number of cities and towns Zhang et al., 2020). For example, Guangzhou and Foshan are becoming more integrated because of transport links (Ye, 2014). In the YRD, foreign investment has transformed the 'Sunan model' (southern Jiangsu) of the collective economy into new spaces of globalisation (Wei et al., 2009), treating the region as an entity and thus creating a regional development model. In the PRD, advanced producer service networks have been formed (Yang, 2020;Yeh et al., 2015a. China has seen rising city-regions Wu, 2012, 2018;Wu, 2016;Lim, 2019). This new spatial form is attributed to different dynamics, for example, foreign investment and changing production networks and extensive knowledge networks (Li and Phelps, 2018;Wei et al., 2009;Yang, 2014;Yeh et al., 2015aYeh et al., , 2015b, inter-city competition (Xu and Yeh, 2005;Zhang, 2006;Zhang and Wu, 2006), increasing transport connectivity Ye, 2014) and governance changes through 'state spatial selectivity', a 'new state space' and state rescaling (Lim, 2019;Sun and Chan, 2017;Wu, 2016). The YRD and PRD have been studied most (Li and Wu, 2018; as well as new national development areas such as Chongqing and Zhuhai (Lim, 2019). Wu (2016) understands China's regional governance through state rescaling. The initial market reform decentralised governance from national to city spaces. However, vicious inter-city competition led to a series of social and environmental problems. The central government then upscaled governance from individual cities to the city region to achieve more coordinated development through spatial planning and policies (Harrison and Gu, 2021;Wu, 2015). In other words, regional governance should be understood alongside state development and governance strategies and concrete geopolitics in the region (Li and Jonas, 2019;Li and Wu, 2020) and the part-state's attempt to regulate its territory through jurisdictional control according to administrative ranks (Cartier, 2015).
New regional development strategies in Jing-Jin-Ji and the Greater Bay Area of Guangdong, Hong Kong and Macau represent state-centred regional governance. The former is seen as a critical step to solve Beijing's environmental crisis, especially air pollution, while the latter aims to mobilise existing cities in the region for cross-border collaboration. The development strategy is to a lesser extent attributed to politics within localities as in western democratic societies (Jonas et al., 2014), owing to the strengthened role of the central government. Collaborations between cities are seen as helpful to enhance integration and achieve the vision of regional development (Yang et al., 2021). Through these new regional developments, this paper shows that current scholarship on China's urban governance generates a wider understanding of the complex politics of development which provides an appropriate characterisation of actually existing megaurban regions beyond market-centred descriptions Wu, 2020a).
Chinese city-regionalism is a scalar fix to combat fierce inter-city competition (Wu, 2016). Hence, we need to interrogate the intentional politics of the state and its governance strategies. These strategies may include multi-scalar state collaborations (Xian et al., 2015;Yang et al., 2021;Zhang et al., 2020, deployment of a new discourse of polycentric urban structure  and coordinated development to achieve a national development vision. Inter-city competition led to broken roads and an unconnected transport infrastructure. Besides being an economic development strategy, the polycentric metropolitan structure is used to address these practical problems of infrastructure coordination (Yang et al., 2021) and growth pressure and climate change. The discourse of polycentricity reflects the multi-scalar and competing power of stakeholders . The idea of polycentricity in China comes from planning input (Harrison and Gu, 2021) and the role of the central government (Dong and Kübler, 2021) rather than being an outcome of 'market' forces as may be seen in enclave urbanism in the West. Through multi-scalar coordination, especially orchestration by the central government, regional development and coordination plans are formulated (Wu, 2015(Wu, , 2016Harrison and Gu, 2021) and negotiations across multi-scalar stakeholders are part of city-region building processes (Yang et al., 2021).
City-region governance in China reveals both economic dynamics (agglomeration and networks) in the regions surrounding the core cities and the politics of regulation and state strategies. Compared with relatively stable local government boundaries in Europe and fragmented government jurisdictions in North America (Cochrane, 2007;Cox, 2016;Keil, 2000), China has seen constant adjustment of administrative boundaries by the central government (Cartier, 2015;Li and Wu, 2012;Ye, 2018;Zhang and Wu, 2006). Indeed, region-building may be directly attributed to jurisdictional changes purposely envisaged by the state selecting the city-region as a coordination space (Wu, 2016). Recent development strategies present a regional vision of mega-cities , for example, the Greater Bay Area or Jing-Jin-Ji, demonstrating distinctive state intervention. In Jing-Jin-Ji, the aim is to relocate economic activities from central Beijing to the cities in the city-region so as to alleviate Beijing's growth pressure. The development of the national new district of Xiong'an goes beyond economic clustering and reveals an attempt to use a polycentric city-region to deal with environmental challenges (Zou and Zhao, 2018). The flagship of Xiong'an reflects not only the role of the central government but also the leadership of the Party. The research thus enriches our understanding of city-region formation in terms of both economic and geopolitical dynamics, especially state politics across geographical scales (Jonas, 2020a;Wu, 2020b). In the Western political system, the 'politics of redistribution' (Jonas and Moisio, 2018;Harrison and Hoyler, 2015) is a key driving force behind cityregionalism, involving mobilisation of different political jurisdictions. In the Chinese case, we have seen a strong role of the central government as well as active local states in region-building. To a lesser extent, the development of city-regions is driven by society demanding regional infrastructure for labour reproduction and inter-city commuting. Beyond economic forces, the concept of mega-urban regions in China has been used as a new imaginary to achieve state development goals (Wu, 2020a;.

VI Conclusion: Implications for urban theory and sustainable urban agenda
There are debates in Urban Studies about universalism and particularism. Brenner and Schmid (2015: 151) argue for a new epistemology to challenge the urban 'as a fixed, bounded and universally generalisable settlement type'. While Peck (2015: 160) appeals for 'a constructive dialogue across theoretical traditions, notably at the interface between political economy and postcolonialism', Storper and Scott (2016) criticise both postcolonial and politicaleconomic approaches as either limited by particularity or by the lack of understanding of general economic processes. The notion of 'planetary urbanisation' (Brenner and Schmid 2014) raises interests for understanding the global South in terms of the connected urban processes in advanced capitalism. From the global South, theories beyond the West are appealed for (Edensor and Jayne, 2012;Parnell and Oldfield, 2014). The perspective of comparative urbanism argues that ordinary cities are not instances or variants of a Western model (Robinson, 2006). They are not created by 'general' political economic processes but rather through 'worlding practices' (Roy and Ong, 2011), for example, through modelling and inter-referencing and learning from the world. Robinson (2016) criticises the separation between the urban context and 'general processes', arguing that contextual importance is beyond particularity. The urban process is exactly transcalar, embedded rather than detached from locality (Robinson, 2016). This paper pays attention to the particularity of the Chinese state (and the specific process of territorial governance), but the focus on China allows us to see simultaneously a broad range of entrepreneurial governance and state centrality, hence creating new narratives beyond the economiccentred imagination (Wu, 2020a). Rethinking Chinese urban 'entrepreneurial' governance reveals the critical role of the state across multiple scales of neighbourhoods, cities and city-regions. While the role of the state in economic and social governance in advanced capitalism is not new (Jessop, 2006(Jessop, , 2016Le Galès, 2016) and recently receives additional attentions (such as new municipalism and statecraft, Lauermann, 2018;Pike et al., 2019;Whiteside, 2021), we begin to understand the more specific role of the state in financialisation and urban redevelopment in China Wu, 2021;Wu et al., 2022).
Using a broad concept of urban governance, this paper aims to theorise the existing scholarship on China by investigating the role of the state in the territories of neighbourhoods, cities and regions. Echoing Friedmann's (2005: xvi) argument that 'stresses urbanization as an evolutionary process that is driven from within, as a form of endogenous development', we try to understand the emergence of Chinese cities in their own contexts in response to their challenges, crises and contests. This also resonates with the appeal of 'grounding urban theory' to precisely pinpoint the context of theorisation (Phelps, 2021;Wu, 2020a). We compare prevailing theoretical explanations about city and regional governance and observations of concrete Chinese forms of governance. By stressing the role of the Chinese state and urban policies, these comparisons aim to go beyond an application of Western theories so as to provide a contrasting or alternative explanation. For example, this paper does not apply the concept of suburbanisation to major peripheral development projects such as the city of Xiong'an outside the capital Beijing (Zou and Zhao, 2018) or Shanghai's Chongming Island for ecological development (Xie et al., 2019). From the perspective of the state, the development of Xiong'an reflects President Xi Jinping's intention for a new development model that is not dependent upon land-finance. Chongming Eological Island indicates multi-scalar governance in which the municipal government of Shanghai aligns its plan with the political agenda of 'ecological civilisation'. Instead, the paper reveals the extent to which developments reflect strategic interventions from the central and local states. The understanding of the role of the state in these mega-urban projects enriches the knowledge of 'global suburbanism' and governance modalities (Ekers et al., 2012). These are extended urbanisations within which transcalar actors interact (Keil, 2018;Robinson et al., 2021). In China, the key process is the reterritorialization of the state.
This paper shows that the state maintains strong strategic input in the process of governance, known as 'state entrepreneurialism' (Wu, 2018b) (Table 1). China's urban governance, as shown in neighbourhood governance, deploys and manages homeowners' associations in middle-class housing estates (Cai and He, 2021) and has strengthened the infrastructure of governance (Tang, 2020;Wu, 2018a). Faced with a fast changing society and migrant urbanism, the state continues to play a key role as shown in urban regeneration and policies to 'incorporate migrants' into 'new urbanisation'. In urban development, the state uses 'market actors'urban development corporations (Feng et al., 2021;Jiang and Waley, 2020) to organise large-scale urban development. Major developments such as planned new towns reveal transcalar state governance (Robinson et al., 2021;Shen et al., 2020). The local state performs the 'ecological fix'  to deal with increasing environmental challenges. In regional governance, city-regionalism is a scalar fix (Wu, 2016) and thus region building is simultaneously state building, namely reconfiguring a state space across cities in a wider region. Concepts such as Jing-Jin-Ji, YRD and the Greater Bay Area present state power and vision and geopolitical intention. The tools of city clusters, partnerships and networks are used to achieve geopolitical objectives (Li and Jonas, 2019;Wu, 2020b), for example, reducing Beijing's 'congestion' and achieving a more balanced regional development. This paper contextualises Chinese urban governance in its historical and endogenous processes. The role of the state in neighbourhoods, cities and regions is understood in the policy and development contexts. We see governance change as a concrete institutional and policy response to existing crises and perceived challenges (see Table 1). At the initial stage of market reform, attracting foreign direct investment was a priority (Chien and Gordon, 2008;Chien and Woodworth, 2018;Wei et al., 2009;Wu, 2015;Yang, 2020). But this governance approach subsequently evolved along with the rising crises of the 'entrepreneurial' world-factory model, the Global Financial Crisis and the changing international political environment (Wu, 2021). Governance has become more state-orchestrated through proactive planning (Wu, 2015) and financial intensification (Wu, 2021;Wu et al., 2020). These governance changes are made, first, to address social challenges. Millions of rural migrants have crossed over the urbanrural dualism and formed social networks within and beyond migrant enclaves. Yet they face various forms of institutional exclusion from accessing public services. The integration of migrants imposes challenges for China's social sustainability. In 2012, the central government proposed the New Urbanisation Plan (2014-2020) to promote people-oriented urbanisation and encourage migrant integration with the host society. The existing literature has extensively documented the social exclusion of migrants (Buckingham, 2017), similar to the situation in the post-welfare state in the West (Peck, 2001). But recent migrant policy changes have not been fully reflected on for what they reveal regarding governance and the continuing role of the state in managing migrant urbanism.
Second, rapid urban growth requires the state to cope with environmental pressure and ecological crises (Chung et al., 2018;Flynn and Yu, 2020;Tian et al., 2015;Zhang and Wu, 2021). Examining the actual national and local politics of governing environmental sustainability reveals both top-down, low-carbon state control and bottom-up, middleclass desire for a better quality of urban living. China's environmental governance goes beyond the 'sustainability fix' that fixes the conditions for capital accumulation (While et al., 2004). Again, similar to enhancing the social governance of migrant urbanism, the central state has strengthened environmental governance. Under President Xi Jinping, a new vision of 'ecological civilisation' is proposed. Regarding the strengthened role of the state in  (2020) suggest that the protean environmental state needs to consider conflicting economic and environmental interests depending upon timing and context. Li and Shapiro (2020) suggest that a new authoritarian environmentalism has been formed. Under multi-scalar state entrepreneurialism, the local state performs the 'ecological fix' to remove low efficiency land uses and reorient towards ecological development, in response to the increasingly strong steer from the central government .
In terms of policy implications, we argue that a subtle and accurate understanding of China's urban governance will help address economic, social and environmental sustainability. Economically, China is facing greater financial risks due to land-finance and local government debts under entrepreneurial development. Socially, the state policy of migrant integration conflicts with entrepreneurial local government which grants urban hukou only to elite migrants. Environmentally, green projects such as 'greenways' in Guangdong are difficult to implement in land development (Chung et al., 2018), despite both top-down carbon control by the central state and bottom-up pressure from the middle-class demand for a better environment (Lin and Kao, 2020). But there are new trends of governance beyond entrepreneurialism . Based on the sound understanding of multi-scalar governance, future research will help to explain and predict the changing direction of entrepreneurial governance faced by rising social agencies on the one hand and state strategies on the other. The understanding of statecentred multi-scalar governance will help to identify appropriate pathways to establish and implement China's new urban agenda.

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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research has received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (grant agreement No 832845, Advanced Grant) ChinaUrban.