(De-)Activating the growth machine for redevelopment: The case of Liede urban village in Guangzhou

Introduction

After more than three decades of rapid urbanisation (Ding, 2007; Han, 2010), many Chinese metropolises are facing an increasing shortage of developable land. As a consequence, the central government has begun to shift its policy focus from fresh greenfield land to land within the existing built-up areas, and encourage local governments to engage in urban redevelopment and improve land use efficiency (Lin, 2015). At the frontier of China’s economic reform, Guangdong Province has thus been entrusted with a new policy experiment, i.e. ‘Sanjiu Gaizao’ (or the ‘Three-Old’ Redevelopment Policy) in 2008. The policy aims at implementing a set of arrangements for redeveloping three different types of old areas – dilapidated neighbourhoods, old factories and urban villages (or chengzhongcun) – across the province.

Of the three types of old areas, urban village is perhaps the most important and intricate one. In the Chinese context, an ‘urban village’ refers to the kind of village that has been spatially encircled by urban development yet still retains an institutionally recognised rural status because of the collective ownership of land (Li and Xu, 2008; Tian, 2008). Existing literature suggests that a ‘win–win’ benefits distribution among stakeholders is critical to accelerating the redevelopment of urban villages (Li, 2002; Li et al., 2014; Tian, 2008). The stakeholders here generally include indigenous villagers, commercial developers and city governments, each with their own set of interests. According to various research, ‘win–win’ benefits distribution among stakeholders is usually achievable through either a ‘collaborative coalition’ (Hin and Xin, 2011), ‘collaborative mode’ (Li et al., 2014), or ‘partnership’ (Lin and Meulder, 2012; Zhao and Webster, 2011). However, largely because of the lack of empirical data, most previous studies seem to have focused on either describing the implementation of redevelopment plans for certain urban villages, or just listing the types and quantities of stakeholders’ interests at a specific time. A detailed analysis of the underlying power relations between stakeholders in the redevelopment process, therefore, is often missing.

On the theoretical front, although much research has examined China’s urban renewal/redevelopment taking place on state-owned land (He and Wu, 2005; Shin, 2009), fewer studies have attempted to extend relevant theoretical models or perspectives to understand urban village redevelopment. Built on previous work on the political economy of urban villages (Hao et al., 2011; Hsing, 2010; Zhang, 2011; Zhao and Webster, 2011), this paper will revisit the typology of place entrepreneurs in the growth machine thesis and explore through a case study how land-based interests embedded in an imbalanced power relationship can (de-)activate urban village redevelopment. Drawing on the data obtained from local governments, this research also intends to set the redevelopment of one particular urban village, namely Liede Village in central Guangzhou, in a wider urban policy context.

The remainder of this paper consists of five sections. The second section will review the literature on one core aspect of the growth machine thesis, i.e. the role of place entrepreneurs in local growth across varied political-economic contexts. The following section will focus on how the idea of growth machine may be deployed for understanding redevelopment activities in Chinese cities. By proposing a revised framework for deciphering the land-based interests of different place entrepreneurs during urban village redevelopment, the fourth section intends to cast the growth machine model in a new light. Set in a changing urban policy agenda, the fifth section presents the case study of Liede Village in central Guangzhou as well as the reason behind city-wide halt of urban village redevelopment afterwards. To sum up the research, further thoughts are offered in the last section.

Place entrepreneurs and the contextualisation of growth machine thesis

In probing how places get built in the USA, the original thesis put forward by Molotch (1976), and later jointly by Logan and Molotch (1987), posits the functioning of local growth machines wheeled by place entrepreneurs for augmenting parochial interests through land-use intensification. According to Logan and Molotch (1987: 29), place entrepreneurs are defined as ‘the people directly involved in the exchange of places and collection of rents’. Three types of place entrepreneurs, or ‘modern urban rentiers’, are key to the understanding of a growth machine, i.e. serendipitous entrepreneurs, active entrepreneurs and structural speculators. Of the three, though, emphasis has been placed on the last type, which is perceived to be capable of altering the prospect and spatial pattern of local development through ‘organisational manipulations’. Despite their different influences, all place entrepreneurs are believed to be gearing towards local growth – which is seen as ‘an aggregate of land-based interests’ (Molotch, 1976: 310).

During the following two decades, while cases studies and comparative research along similar lines had abounded (Jonas and Wilson, 1999; Logan et al., 1997), debate on the composition and operation of growth machines/coalitions in different contexts appeared to have relegated place entrepreneurs to a less prominent position. As Rodgers (2009: 16) has noted, ‘the complex sets of agency supposedly making up growth coalitions remain in empirical practice a largely unexamined, independent variable’. This said, among those who do accord importance to place entrepreneurs, Molotch (1979) has reiterated the intermediary role played by landed elites in ‘preparing the ground’ for local growth, with the caveat that the fine line between parochial rentier and non-local capitalist would begin to erode. Elsewhere, urban political-economic conditions specific to Japan and Italy have been illustrated by case studies, in which local place entrepreneurs are found to be constrained either by the hierarchical structure of the state, or the in situ political/party system (Molotch and Vicari, 1988; Vicari and Molotch, 1990). However, apart from a number of studies which have touched on the contextual differences between British and North American growth machines (Ward, 2000; Wood, 1999), research attention is yet to be adequately paid to place entrepreneurs in more non-US settings (Harding, 1999; Rodgers, 2009).

Over the past 10 years, there has been renewed interest in retooling the growth machine thesis for urban research (Light, 2002; Phelps, 2012). More importantly, increased attention has been paid to a few post-socialist countries in an attempt to engage Western theoretical model with local realities (Golubchikov and Phelps, 2011; Kulcsar and Domokos, 2005). In a rather different context, Kulcsar and Domokos (2005: 551) have re-confirmed that ‘the growth machine is essentially a power elite model’ (see also Harding, 1995). However, given the specificities of social and political-economic situations, the functioning of growth machine will vary considerably – not only in the organisation of place-bound elites, but also the ways in which external investors and state resources are orchestrated locally. Mapping the placeless growth of Khimki, a municipality on the edge of Moscow, Golubchikov and Phelps (2011) have reported the development of a post-socialist suburb in a nested political setting – which features a lack of coordinated elite/coalition in a compressed urbanisation process. The agency of land-based interests notwithstanding, it seems that the political-economic conditions under which place entrepreneurs operate are also essential to explaining the mechanism of a growth machine. In more recent studies, the indispensable role played by the state, local or otherwise, has as well been highlighted in the cases of reshaping Tysons Corner (Phelps, 2012) and Houston’s Chinatowns (Knapp and Vojnovic, 2013) in the USA. To bring the above discussions into conversation with relevant work on a fast-urbanising China, the next section will examine how the idea of growth machine/coalition may inform cross-national research in another distinct context.

Locating (pro-)growth machine/coaltion in urban China

Over several decades, the growth machine theory has been an influential model for explaining city development and urban renewal in the USA (and beyond) (Gotham, 2001a, 2001b; Jonas and Wilson, 1999; Logan and Molotch, 1987). As the evolvement of growth machine thesis has in large measure been accompanied by the rising popularity of regime analysis in urban politics (Harding, 1999; Logan et al., 1997), quite often the notion of ‘pro-growth coalition’ would be adopted to characterise an associated group of elites who apparently dominate the prospects of a locality. For those who may want to distinguish between these two approaches to urban analysis, i.e. growth machine and urban regime, some difficulties would normally arise. More often than not, the bonding mechanism of local landed elites has tended to be underplayed in most case studies, whereas the political side of urban growth would usually eclipse the part played by land and property in coalitional analyses.

So far, research into the working of (pro-)growth coalitions in China, with few exceptions, has also been focused on the political dimension of urban (re)development. Even though great progress has been made in reforming the country’s economic realm since the late 1970s, urban (re)development in Chinese cities is still very much a state-dominated process (Wu, 2002, 2016; Wu et al., 2013). Economic growth, a top priority on central and local governments’ agenda, has to be enacted by politically entrusted territorial units, whether urban or rural, through a hierarchical state. In a rescaled context where urban landed interests (rural land and properties excluded) have become marketable and thus essential to economic development, local (pro-) growth coalitions have proliferated; and the growth machine thesis, right because of its focus on land and property development, is arguably a very productive tool for understanding a rapidly urbanising China, where city expansion, suburbanisation and restless redevelopment inextricably overlap.

While there have been earlier attempts to localise the idea of (pro-)growth coalition in a few major Chinese cities (Zhang, 2002; Zhang and Wu, 2008; Zhu, 1999), other studies have suggested that certain features of urban growth machine are well to be found in China’s dynamic urban expansion and redevelopment (Han, 2010; Zhang and Fang, 2004). Of late, a body of literature has emerged in an effort to approach urban transformations in China from a political-economic perspective (Hsing, 2010; Lin, 2007; Wu, 2015). Most notably, urban coalition and (pro-)growth machine are among the revisited concepts thought to be useful for deciphering land-centred redevelopment politics across the country (Yang and Chang, 2007; Zhang, 2014). Contrasting the mechanisms of urban growth in inner-city areas or on newly-developed land, battles fought on collective turfs, i.e. redevelopment of (sub-)urban villages, have also taken scholarly interest beyond the boundaries of state-owned land (Lang et al., 2016; Wu, et al., 2013, 2016).

As a significant component in the redevelopment process, urban villages represent a peculiar type of space located at the urban and rural interface, posing both theoretical and practical challenges to policy-makers and researchers in many Chinese cities. The conundrum of urban village redevelopment stems from the duality of China’s land management system (Lin and Ho, 2005), whereby villages encircled by rapid urban expansion have become collectively owned enclaves amid state land. So far, although much debate has pointed out the institutional obstacles to the redevelopment of urban village (Lai et al., 2017; Liu et al., 2012a; Song, 2015), localised configurations of landed interests per se still call for more research either from a political-economic vantage point (Zhang, 2011), or a more pronounced coalitional approach (Hin and Xin, 2011; Lin et al., 2015).

It is argued that not only in physical, but also in institutional terms, redevelopment of urban villages should fit aptly into a place-bound growth process which embodies some key features of Logan and Molotch’s theorisation (1987). Most importantly, the dualistic land/property ownership system underlying the interests redistribution during urban village redevelopment has in fact revealed an imbalanced power relationship between a few basic types of place entrepreneurs in urban China. Furthermore, given the salience of land and property in urban village redevelopment – most often in the forms of on-site or relocated re-settlement and monetary compensation, the political economy of reshaping such transitional spaces could also enrich our understanding of locality growth in a rather different urbanisation context. However, to better tailor the growth machine thesis for cross-national utilisation, we are also aware that its original theoretical model will need to undergo some revision before empirical analysis could proceed for a fruitful dialogue (Jonas and Wilson, 1999).

Growth machine and urban village redevelopment: A revised framework

A redevelopment growth machine working in the grounds of China’s urban villages needs to source its conceptual elements locally. Yet, certain component parts originally offered by Logan and Molotch (1987) for profiling a growth machine in the USA are still highly pertinent. Despite institutional dissimilarities, a social typology of place entrepreneurs is readily distinguishable in China’s urban villages. First, while ordinary urban villagers are like a proactive group of serendipitous entrepreneurs, the village collective – community orgainisation representing the overall interest of an urban village – could well be seen as an in situ ‘structural entrepreneur’. All of them, needless to say, rely heavily on the rents derived from the locational benefits of their village, especially when it happens to be near the city centre. The village collective, in particular, with its representative bargaining power, is normally comprised of indigenous elites able to negotiate with or even contend against outside developer and local government when defending community interests.

Second, unlike in a Western growth coalition, local governments entrusted to monopolise de facto public land ownership in Chinese cities are a distinctive type of agency. As far as redevelopment of urban villages is concerned, local governments acting as proxy state-land owners – be they at the county, district, city or municipal level – that have a final say not only on the development of public land, but also on the ascertainment and urbanisation of rural land within their administrative confines. In contrast to urban villagers and their collective organisation, who tend to stick to parochial interests, local governments normally would have broader visions and policy objectives for urban (re)development across their territorial area. More to the point, under a dualistic land/property ownership system, political powers over legally acquiring, converting and then transacting rural land on the market to make it ‘urban’ are a monopoly of specific levels of government (see the case study below). Although with many characteristics of a structural speculator as depicted in Logan and Molotch’s original thesis, it is the overarching decision-making power that ultimately defines the agency of China’s local state. Rather than intermediating between rentiers and developers, local governments here are far more authoritative and powerful in enabling locality growth and managing land-use change (He and Wu, 2005; Lin and Ho, 2005). And indeed, we would argue that most (pro-)growth coalitions formed as such will hinge upon local governments acting as a ‘monopolistic land manager and place entrepreneur’, who would naturally become the pivot of growth machine politics.

In addition, the functioning of a growth machine/coalition for urban village redevelopment follows a more complicated logic than that of ‘value-free’ development (Logan and Molotch, 1987). To leverage the financial resources from real estate developers, such machinery has to secure the involvement of market forces with government blessing or endorsement. Typically, urban village redevelopment is guided by local governments towards a growth agenda, often through property-led renewal (Hin and Xin, 2011; Wu, 2015). With rules of game set and locality-specific goals formulated, it is subsequently for those active entrepreneurs to strategise on the prospects of redevelopment investments. In fact, fiscally tight local government would rather encourage an urban village and developer(s) to form a self-administered coalition than have to incur financial burden on public coffers. In this sense, we would argue that the original growth machine model could be complemented by extending the rubric of active entrepreneurs to include certain institutional investors; operating in a circumscribed arena – as outside developers have to navigate the complexities of speculating on converted rural land, these strategic entrepreneurs are much less dominant than those landed or corporate elites going for ‘value-free’ development in a capitalist society.

For a growth machine to work for urban village redevelopment, it is crucial that land-based interests should be generated and redistributed in a way that binds the involved parties together. At the heart of growth machine theory is the dialectics of use value and exchange value of land. In our revised social typology of place entrepreneurs, however, the value scales are arguably inclined more towards ‘exchange’ than ‘use’ – partly because urban village redevelopment entails more than planning and building on a tabula rasa. Although all coalitional parties will expect to maximise their gains, the forms in which their interests take could vary. Anticipated as well as realised growth in land/property price being the ultimate source of value increments, we would propose a more elaborate framework for deciphering the land-based interests involved in activating a growth machine for urban village redevelopment.

What is more, urban village collective and villagers are place-bound actors with landed interests. Both of them have use value to secure during redevelopment, albeit out of differing concerns. Urban villagers losing homestead in the process are primarily concerned about replacement housing and compensation, whereas the village collective, representative of community interests, is due to safeguard and increase the village’s welfare with the whole locality in mind. On top of use value, redevelopment is also expected to bring substantial rent increments (or else there would be no need to have the village bulldozed and rebuilt) – especially for those centrally located urban villages. Perhaps not entirely a windfall, the increased exchange value of redeveloped land/properties is certainly much welcomed, if not actively demanded, by the village collective and villagers. Unlike villagers, who through inheritance are serendipitous individuals, the village collective even has institutional capacity stemming from collective land title to bargain with outside players as well as change the structural conditions for locality development. Of course, a key element to urban village redevelopment is the engagement of real estate developers. To assemble a growth machine able to secure funds from committed developers, the exchange value of land/property after redevelopment will have to be substantial enough to entice strategic investors (most probably) from outside an urban village.

On the other hand, through its monopoly over urban land and administrative powers – particularly planning, local government plays a dominant role in the redevelopment process. Despite unresolved urban–rural cleavage, local government tends to view the city as a territorial ensemble geared towards the accumulation of land-centred interests (Hsing, 2010; Lin, 2007, 2009). In order to realise a tally of political, economic, social, and environmental interests (Chen and Xu, 2011), local governments would usually take the steering position in handling various policy issues, balancing social and environmental concerns, or even making concessions where necessary to either accelerate or slow down redevelopment. As to urban village redevelopment, local government would be more concerned with an effective solution to the issue in general, although they do prioritise certain villages over the others, recognising locational specificity. Given the rural perimeter, it would seem that neither the use value nor the exchange value on collective land may be directly gleaned by local government; nevertheless, villages thriving within the city do cause headaches for urban management. Issues of public safety, sanitary regulation, migrant population and shanty images … could cause surrounding state-land to lose part of exchange as well as use value. Local government, as ‘monopolistic land manager and place entrepreneur’, will likewise have considerable interests at stake.

By linking conceptual framework with empirical research, the remainder of this paper will present the case study of Liede urban village in Guangzhou, a metropolis in south China, where the activation and halting of redevelopment growth machine has plenty to offer for comparative thinking.

The (de-)activation of redevelopment growth machine: A case study of Liede urban village

In Guangzhou, the provincial capital of Guangdong and the largest city in south China, because of rapid urbanisation over recent decades, a growing demand for both low-rent informal housing by rural migrants (Wang et al., 2009) and centrally located commercial and office space by enterprises has led to a proliferation of urban villages in the city. In 2009, altogether there were 138 officially registered urban villages across Guangzhou (according to ‘the 56th policy’, see below). As early as the 1990s, the city government had sought to redevelop some of the villages located in the inner-city area, but not until the year 2010 did it finally succeed in redeveloping the village of Liede. The first of its kind and probably the most representative case (Sun et al., 2017), Liede village offers an important opportunity for us to look into the activation of a growth machine for urban village redevelopment.

Motivated by the ‘Sanjiu Gaizao’ Policy, the Guangzhou government announced an ambitious scheme in 2009 to completely redevelop 52 urban villages over a period of 3 to 5 years. The scheme was publicised in a policy document entitled ‘Opinions on Accelerating the Redevelopment of Old Towns, Old Factories and Urban Villages’ (commonly known as ‘the 56th policy’). By April 2012, redevelopment plans for more than 20 urban villages had been approved by the city government, of which nine centrally located ones were to be demolished before the 2010 Asian Games (GURB, 2012b).

Liede village occupies the south and central part of Pearl River New Town (PRNT), the new central business district of Guangzhou, and is situated on the bank north of the Pearl River, across from Haixinsha Island (Figure 1), where the opening ceremony of the 16th Asian Games was staged in 2010. Before redevelopment, it had 3167 local households (totaling 7865 villagers) and about 17,000 migrants within an area of 336,000 m² (GURB, 2012a).

OPEN IN VIEWER

Figure 1.

Location of Liede village in the PRNT and Guangzhou.

The pre-redevelopment context: City government versus the Liede Village Committee

In the late 1990s, in the wake of possibly losing its leading position in regional development, the city government of Guangzhou proposed a long-term strategy to improve its physical environment so as to become an international metropolis. Urban village redevelopment was therefore brought onto the government’s working agenda. However, when speaking on city-wide visions, the then mayor of Guangzhou made it clear that developing new districts should take priority over redeveloping urban villages – for newly developed districts were expected to reduce the resident population in urban villages and lower the cost of subsequent redevelopment (Lin, 2013). In other words, the policy objective then set for existing urban villages was to improve their physical environment and decrease urban density.

One effective approach to achieving the objective was to control the redevelopment FAR. In 2004, the city government issued some guidance for Liede village redevelopment in an official document, the main points of which included: (1) the total redevelopment FAR must be lower than 4.0; (2) on-site resettlment must be done according to the principle of ‘one household one plot’; (3) the land area of all rebuilt houses must each be smaller than 100 m² and no higher than 3.5 storeys (GUPB, 2004). Following these instructions, the ‘Redevelopment Plan of Liede village in Tianhe District’ (2005) then set the net redevelopment FAR at 3.95, the gross residential floorspace at 665,000 m² and the collective-owned property (a planned hotel) at 120,000 m² (GTDUPB, 2005) (Table 1).

OPEN IN VIEWER

Table 1.

Comparison of redevelopment figures for Liede village.

Components	Redevelopment plan in 2005	Final redevelopment plan
Overall floorspace (m2)	916,000	1,421,000
Residential floorspace (m2)	665,000	687,000
Construction floorspace of collective properties (m2)	120,000	174,000
Net FAR	3.95	5.52 (including auctioned land plot)

Sources: GTDUPB (2005); GUPB (2007); GURB (2012a).

In December 2005, the Liede Village Committee refused to accpet the plan and made representations to the government. Their arguments mainly focused on land-based interests: (1) compared with surrounding urban plots, the 4.0 FAR was too low and hence the use value of redeveloped site would not be fully realised – and their opinion was to raise it to 6.0–6.5; (2) the FAR set for the collective-owned properties was also considered too low and should be revised; (3) besides some amenities concerns, suggestions were made regarding a residential plot in the east of the redevelopment site, the idea of which was to redistribute the development rights thereupon into other residential plots by further increasing the planned FAR (Liede Village Committee, 2005a).

As an important side note, in order to protect public interest, the city government had excluded real estate developers from taking part in Liede village redevelopment since 1999. This was done through the government’s monopoly over the primary land market, with the assumption that developers might push up the redvelopment FAR or cause more social conflicts because of insufficient compensation – both of which had indeed occurred in some urban renewal projects in Guangzhou during the mid-1990s (Lin, 2013). Regarding the requisite funds for urban village redevelopment, the government spelt out three principles, i.e. ‘no governmental investment’, ‘no sourcing from real estate developments’, and ‘those who would benefit the most shall invest’ (Yan et al., 2004). These principles were essentially saying that it was the villagers who should provide funds for any redevelopment, while the city government, as the monopolistic owner of urban land around the village, was merely to take a containment attitude towards the negative externalities of future redevelopment activities in an urban village. Without ‘eminent’ needs for intervention, the city government would rather leave urban villages to their own devices than step onto the collective turf by taking substantive measures.

Redevelopment growth machine deactivated

Since May 2012, the city government of Guangzhou has approved no more urban village redevelopment plans. The previously announced city-wide redevelopment scheme targeting 52 urban villages has likewise been suspended. The reasons behind this decision, as the following analysis reveals, must be understood in a wider policy context.

First, perhaps quite unexpected by the city government, large-scale urban village redevelopments had turned out to be a potential threat to its land-leasing revenues. Up until the 16th Asian Games, the city government had raised a considerable amount of construction funds by establishing government-owned investment vehicles to mortgage urban land for bank loans; and the resulting urgency in debt repayments compelled the government to continue acquiring and leasing more land in the urban fringe areas through further developing new towns. Meanwhile, macro regulations over the national real estate market had curbed demand for urban construction land, causing a steady decline in the transaction volume. What Figures 2 and 3 show is the diminishing ratio of trading quantity to supply quantity of residential and commercial land, and a corresponding decrease in annual land-leasing revenue. The Guangzhou city government was quick to place the blame on the already approved urban village redevelopment plans – which were to supply 11.9 million m² floorspace of new developments on collective land in the inner city (GURB, 2012b) – for impacting on the demand for land in fringe areas and causing the drop in land-leasing revenue.

OPEN IN VIEWER

Figure 2.

Ratio of trading quantity to supply quantity of residential and commercial land in Guangzhou, 2002 to 2011.

OPEN IN VIEWER

Figure 3.

Annual land-leasing revenue (billion RMB) of urban residential and commercial land in Guangzhou, 2002 to 2011.

Second, ‘the 56th Policy’ of 2009 on ‘Sanjiu Gaizao’ had included a principle encouraging ‘self-initiated redevelopment and transaction of commercial-use plots through closed-door negotiation’. Considering this kind of transaction was least transparent and most easily manipulated, the city government feared that it would direct more developers away from competitive transaction of urban village land and lead to a decrease in land-leasing premiums for the government’s extra-budgetary income. As a result, from June 2012, ‘the 56th Policy’ has been superseded by a new policy document, which in effect has strengthened governmental control over urban village redevelopment.

Furthermore, large-scale urban village redevelopments also posed a serious problem for the city government to manage the built environment. Before redevelopment, urban villages at central locations were already very dense – for instance, the original gross FARs of Xiancun village, Yangji village and Linhe village had all exceeded 3.0. The approved redevelopment FARs for the former two are understood to be well above 6.0 (GURB, 2012b), much higher than that of Liede. This upward spiral hardly bodes well for the city government’s planning control. In addition, under the 56th Policy, there was much room for villagers and developers to possibly ‘collaborate’ on the proposed redevelopment FARs and misrepresent the actual cost of redevelopment for getting as much compensation from the process. Left with little say in ascertaining the redevelopment FARs, the city government felt that it was simply unable to monitor either the total floorspace of in-kind compensation or the cost of redevelopment.

And such are the underlying reasons for a new leadership of the Guangzhou government to have held up urban village redevelopment. The latest progress since has included: closed-door negotiations for land-leasing in urban villages were tightened up and replaced by either public tender, auction, or listing – all more transparent and competitive ways to increase the government’s land-leasing income; as regards the redevelopment FARs being pushed up by exaggerated cost of redevelopment, the city government has stepped up measures to monitor the actual costs by carrying out the work with relevant village collectives, while excluding developers from such verification processes.

Discussion and conclusion

In recent years, there has been growing interest for local governments in China to engage in redevelopment activities as a result of increasing shortages of urban development land. Against this background, urban village redevelopment has become a heated issue. From a political-economic perspective, when compared with more established theoretical frameworks to comprehend China’s urban growth, redevelopment of urban villages still remains a much-debated yet under-synthesised field of enquiry. By investigating the land-based interests of major coalitional parties through the case study of Liede, this research is intended to explore the (de-)activation of a growth machine for urban village redevelopment in the policy context of Guangzhou.

To bring forward the role of place entrepreneurs in engineering locality growth, a revised typology has been proposed for this study. Accentuating land and property interests and (local) state dominance in redeveloping an urban village, we have identified four (types of) place entrepreneurs as the main parties in a growth coalition, i.e. serendipitous entrepreneurs, active institutional entrepreneurs, structural entrepreneurs, and monopolistic land manager and place entrepreneur. In retrospect, however, as shown by the shifting course of urban village redevelopment in Guangzhou, the power of local government obviously prevails.

The case study shows that the growth machine thesis is very helpful to reveal the complicated stakeholder relations and highlight the dynamic exchange of powers between the different players. It offers a good analytical framework for the study of urban village redevelopment process. As the strongest party in the collation, the city government may activate or de-activate the growth machine in urban village redevelopment, only to rebalance the benefits shared by different parties. Decision on Liede Village has been influenced by the Asian Games, which reflected the market conditions at the time. Indeed, urban village redevelopment in all Chinese cities require a particular market and economic condition. When the economic growth is at a high speed and property market inflation is high, urban village redevelopment is easier; otherwise, there will be less incentive for activating the growth machine. While the Asian Games provided a momentum for economic development at a particular point for Guangzhou, other events could have a similar effect at a different time and location. Given the general pattern of power structure, the detailed practice of redevelopment indeed may vary.

Rooted in a dichotomous system, the collective ownership of village land provides an important variation on the operation of growth machine in China. With many legal ambiguities of the land ownership, collective land could be interpreted and treated very differently in different places even within the same city. The Liede case here shows that a stronger, active and united village collective organisation – the village committee, will be essential for the successful and timely operation of the growth machine which brings favourable economic benefits to the households who shared the ownership of the land. On one hand, urban villagers have fought hard for both the use and exchange values to be reaped from serendipitous land and property after redevelopment. On the other hand, the village collective, when negotiating on behalf of the locality, are structurally parochial not only in physical scale, but also in the scope of power; its manoeuvring is centred on further densification of the built environment for rent-seeking, but will only be effective to a certain extent. When facing local government’s decision-making power, the collective has to yield to a higher-level authority bearing down on local fortunes. Nonetheless, the benefits local government can gain from such redevelopment would be arguably less about returns in terms of use value or exchange value on collective land (although indirectly it does get some); nor is one particular urban village the only concern for the government to address. As the proxy owner of state land within its administrative area, local government – in this research the Guangzhou city government – has its own land-based interests to pursue. Yet given the territorial mandate to balance as well as prioritise certain (re)development policies, it does have the need to tackle knotty issues such as redeveloping existing urban villages.

In the case of Liede, an endorsing local government, enterprising villagers and village collective, and the financial input of developers are all part and parcel of a unique kind of growth machine. However, the deactivation of similar projects soon afterwards has revealed something more about the underlying logic of local decision-making. It would seem that the city government will have a variety of policy goals to juggle – economic, political, social, etc. – before deciding whether or not to proceed with urban village redevelopment, or indeed, any kind of major development schemes. Normally, urban village redevelopment is high upon policy agenda because it tends to be physical changes with ready visibility. Often a thorny issue, successful redevelopment, especially those at central locations, would usually be viewed as a major achievement for the office term of a certain leadership. However, should there be more immediate concerns emerging, say fiscal constraints, things could also quickly change track. What is more, in a decentralised administrative system, local governments in China are empowered to make certain hard-and-fast rules. Typical instances include making policy exceptions (e.g. through planning), granting/excluding entry of market forces into local projects, and harnessing social participation in urban policy-making, etc. By exploring why Liede village had been successfully rebuilt before the 2010 Asian Games and what reason has led to the halt of similar redevelopment projects in Guangzhou, this research attempts to offer a political-economic approach for explanation.

In the past, the growth machine model has been mainly focused on the local level. This study, however, has attempted to set the case of Liede village in a wider urban policy context. Such an effort by no means seeks to provide a generalised Chinese growth machine model through a one-off project. As another article in this special issue illustrates, the perspective of urban regime may also, on a different theoretical plane as well as spatial scale, be useful for interpreting various redevelopment policies and practices in China. With a comparative view in mind, though, it is hoped that this paper could further engage Western theorisation with Chinese reality, especially by exploiting the potential of urban political-economic perspectives to unravel some challenging issues in a fast-urbanising China. Urban village redevelopment, in this respect, is still in need of more theoretical cross-pollination and inspiration.