Administrative restructuring and urban development in China: Effects of urban administrative level upgrading

Abstract

Chinese

This study focuses on state-led urbanisation in the Chinese context by examining the effects of urban administrative level upgrading on urbanisation and urban development. We argue that urban administrative level upgrading is an important policy instrument in China that empowers cities to play a leading role in driving national and regional economic development. However, there is a lack of systematic empirical investigation to determine whether the objectives of urban administrative level upgrading have been achieved. Using a quasi-experimental method that combines propensity score matching with the difference-in-difference approach, we thoroughly examine the effects of county- to prefecture-level city upgrading and county to county-level city upgrading on the development performance of cities. The results confirm that county- to prefecture-level city upgrading can positively lead to a significant increase in urban population growth and fiscal revenue in a few years after upgrading, although this may not necessarily lead to rapid industrialisation. However, the same is not true for county to county-level city upgrading. This difference may be because counties and county-level cities are at the same administrative level, whilst prefecture-level cities are at a higher administrative level compared with county-level cities and counties. The results indicate that China’s administrative division system, although restructured, retains hierarchical and level-based structure in the post-reform period. We highlight the continuing strategic role of the Chinese state by establishing new scales and arenas and adjusting administrative structures to promote urban development.

Keywords

administrative division system China state-led urbanisation urban administrative level upgrading urban development

Introduction

China has long been an agrarian country in which most of the population resided in the countryside. This situation has completely changed after rapid urbanisation in the past four decades. The urbanisation level increased from 17.9% in 1978 to 56.1% in 2015. In 1978, China only had 193 cities, but this number increased to 656 in 2015. Urbanisation, seen as a spontaneous and self-generated course accompanied by industrialisation, is mainly driven by market forces in a free market economy (Bairoch, 1988; Berry, 1981). In China, however, the role of the state in affecting urbanisation and urban development is critical. Since 1978, market-oriented reforms have led to substantial changes in state–market relations, or more specifically, ‘the Chinese state has essentially externalised the market mechanism’ (Yeh et al., 2015: 2827). Nonetheless, the state has maintained its dominant role, playing ‘a regulatory, interventionist and participatory role’ (Ma, 2002: 1563) in urbanisation and urban development, which differs dramatically from that in advanced capitalist countries. The strong state presence as a key feature of China’s urbanisation is well noted (Ma, 2002; Shen, 2006; Wu, 2002, 2016).

One of the salient features reflecting the omnipresent state is that the state maintains powers to restructure the administrative division system to accomplish geographically or spatially targeted urban development (Cartier, 2013; Chung and Lam, 2004; Liu, 1996; Ma, 2005). All Chinese territorial units are governed administratively through a hierarchical and level-based system, which is structured in four basic administrative levels (xingzheng jibie or dengji), namely, provincial, prefecture, county and township levels. Given that subnational territorial units are not constitutionally guaranteed in China, changing and restructuring the administrative division system has been a common and effective strategy implemented by the central and local governments to achieve development. Aiming to bring the state back to the centre of scholarly inquiries on China’s urbanisation, Cartier (2015) developed the idea of territorial urbanisation as the process through which the state governs the administrative divisions to establish, expand and develop cities.

There has been a strand of literature that focuses on the interface between administrative systems and the economic spaces. These scholars have studied administrative restructuring, such as changes in territorial boundaries, reforms of local governments and reallocation of functional responsibilities, which have occurred in response to economic changes in many countries since the 1960s and 1970s (Bennett, 1997; Maurel, 1989). Globalisation, the rise of neoliberalism and post-Fordism transition have jointly resulted in the reorganisation of administrative space in Western developed countries. Scholars thought that territoriality and geography were being dissolved as capital expanded globally (O’Brien, 1992) and that the traditional administrative boundaries had become irrelevant, redundant or obsolete in contemporary capitalism (Brenner, 1999). From the theoretical perspective of scale, scholars have argued that the state power has been ‘hollowing out’ upward to a supranational regime, downward to regional and local institutions and sideways to non-state sectors (Jessop, 2004). As a result, new scales and arenas have been established and administrative or jurisdictional structures of the state have been adjusted (Jessop, 2002). Many studies have been carried out to explore how the reorganisation of the administrative system and governance structure influences urban and regional development in the global economy (McGee, 1998; McGinnis, 1999; Smith, 2003).

The literature mainly follows the experiences of administrative restructuring in North America and Western Europe. In these countries, territorial units are autonomous entities that are horizontally interconnected as a bottom-up network. Conversely, China has a hierarchical and level-based administrative division system. Thus, China’s case deserves close examination as such a study can be an important contribution to this strand of literature on the administration/economy interface. A growing number of researchers have studied how the state affects urbanisation and urban development through administrative restructuring strategies in China, such as the administrative annexation or territorial merger (Cartier, 2015; Zhang and Wu, 2006), establishment of new districts (Li, 2015), implementation of city-leading counties (Chung, 2007) and making of city-regions (Li et al., 2014; Liu et al., 2012). However, the upgrading of urban administrative level, as a critical administrative restructuring strategy, is relatively under-researched. China’s restructuring of the administrative division system in the post-reform period has been centred on cities, and this strategy is implemented through a form of urban administrative level upgrading that places cities, especially large and medium-sized ones, at high administrative levels within the administrative division system (Ma, 2005). The city-centred administrative restructuring aims to stimulate the development of cities and to facilitate urbanisation. However, there seems to be a lack of systematic empirical investigation to determine whether the objectives of urban administrative level upgrading have been achieved.

Against this background, the present study examines the reorganisation of power relations in the urban administrative level upgrading processes and its effects on urban development performance. In China, a territorial unit’s administrative level strongly determines its political and economic powers and significantly affects its power relations with other places. The urban administrative level upgrading strategy involves an upward shift in scale. The concentration of powers towards cities, which has significantly influenced urban development in post-reform China, deserves closer examination. As argued by Shen (2007), the theoretical perspective of scale should be adopted to understand the reconfigurations of administrative levels and territorial borders that have been undertaken since the early 1980s. We argue that urban administrative level upgrading is one of the most important strategies of territorial urbanisation in the post-reform period. Upgrading a city to a higher level bestows it with stronger powers to play leading roles in regional economic development.

In the following section, we will elaborate how urban administrative level upgrading used as a policy instrument to promote urbanisation and urban development in post-reform China. The operation mechanisms of urban administrative level upgrading and the reorganisation of power relations in the upgrading will also be elaborated. Next, we empirically examine the effects of urban administrative level upgrading on urban development, which is the main purpose of this strategy. Using a quasi-experimental method that combines propensity score matching (PSM) with the difference-in-difference (DID) approach, we will examine the effect of urban administrative level upgrading on the development performance of Chinese cities. This study contributes to the growing literature on state-led urbanisation in post-reform China.

Urban administrative level upgrading as a strategy ofstate-led urbanisation in China

City-centred administrative restructuring in post-reform China

Scholars in general have a consensus that the state plays a critical role in China’s urbanisation process (Ong, 2014; Shen et al., 2002; Wu et al., 2007; Yeh et al., 2015; Zhang, 2008). Thus, to understand urbanisation in China, the ways by which the state intervenes in urban development need to be considered seriously. The use of administrative means to promote economic activities has been characterised as a salient feature in China’s political economy system (Ma, 2005). With a rigid hierarchical and level-based administrative division system, the central government can govern the economy and society by using administrative means (Chan, 2010). In the pre-reform period, with a command economy functioning through the hierarchical administrative division system, the central state has dominantly controlled all territorial units in development plans and political ideology. Local places, such as provinces, prefectures, cities and counties have served as links in the administrative chain to implement centrally designed social and economic development plans (Ma, 2005). Planning quotas and development resources were distributed downward through this hierarchical system. Consequently, local governments were left with little decision-making power for their economic development. Since 1978, China has implemented market-oriented reforms, which have effectively dismantled the command economy system. As a result, the former highly centralised administrative division system has been experiencing restructuring, which is characterised by power devolution and deregulation, and thus, subnational territorial units enjoyed significantly expanded powers to develop their economies (Harvey, 2005).

The administrative restructuring in post-reform China is characterised by city-centred initiatives (Ma, 2005). To implement the market-oriented reforms, the focus of state policies had shifted from the countryside to cities. This development strategy is known as ‘urban-biased’ and aims to achieve the effectiveness of economic growth. Furthermore, this strategy took the cities as the engine of economic development in the post-reform period (Oi, 1993; Yang, 1999; Yang and Chang, 2017). At the fifth session of the Fifth National People’s Congress in 1982, the Chinese central government proposed to create central cities at every level to organise and manage economic activities and stimulate rural economic growth, which was then labelled as the starting point of ‘a central city-based regional planning strategy’ (Kirkby, 1985: 230). Consequently, urban administrative level upgrading has been implemented to place selected cities at high administrative levels and empower them to govern low-level territorial units. The city-centred administrative restructuring involves changes in administrative category, shifts in hierarchical relationships and the revision of territorial boundaries, thus giving rise to a new form of spatiality in China’s territorial organisation, which was historically based on provinces and counties (Chien, 2013; Fitzgerald, 2003; Ma, 2005). A clear tendency in the downward shift of the state power from a single unitary national scale to multiple local scales has been highlighted (He and Wu, 2009; Li and Wu, 2012; Wu, 2002).

Studies have investigated various types of administrative restructuring strategies. Ma (2005) applied the scale theory to inquire into three strategies of administrative restructuring, namely, ‘city administering county’, ‘converting county to city’ and ‘annexation of suburban counties’, with a focus on the reshuffling of the state power in the restructuring. He emphasised the administrative level of a city as a special type of scale that is closely related to its administrative and economic powers. By viewing urban spatial transformation and reconfiguration as rescaling and territorialisation processes, Shen (2007) argued that the central government has substantial power in the process of changing city scale and territorialisation. Zhang and Wu (2006) studied the changes in regional governance in the Yangtze River Delta by studying administrative annexation. They found that restructuring an administrative division system that was incompatible with economic reforms strengthened the intervention of local governments in the economic transition. Luo et al. (2014) examined the ‘province-leading-county’ as a scaling-up strategy based on a case study of Jiangsu, and found intense power struggles among governments at different levels.

Meanwhile, prior studies have provided contextual explanations for administrative restructuring in relation to urbanisation in post-reform China. Cartier (2015) developed the concept of ‘territorial urbanisation’ to characterise targeted changes in the administrative division system, through which the state can achieve the goal of urbanisation and urban development. Similarly, Liu and his co-authors proposed the idea of ‘administrative urbanisation’ (Liu et al., 2012; Yin and Liu, 2017), which indicates that urban development in post-reform China is directly influenced by local state policies that are implemented by strong administrative means. This idea can explain the micro-level state interventions that emphasise the importance of local states and their administrative tools, such as land leasing, investment attraction, new town construction and relocation of government office buildings.

These studies have highlighted the decentralisation and recentralisation of state power among high- and low-level authorities, giving rise to the intra-governmental relations and contradictions of multi-level governance structure. However, most studies are case-based and focus only on place-specific rescaling practices. Very few studies have examined the national-level restructuring of the administrative division system. Moreover, existing studies mainly investigated the underlying mechanisms of administrative restructuring practices but seldom shed light on their effects on urban development performance, which is the main goal of the central government’s restructuring of its administrative division system. The present study attempts to fill these research gaps.

Operation of urban administrative level upgrading

In China, each territorial unit, including the province, prefecture, city, county, league (meng), urban district (shixiaqu) and banner (qi), falls under one particular administrative level. Although restructured, the administrative division system remains hierarchical and vertically linked. The administrative level of a territorial unit significantly affects power in terms of administrative, fiscal and economic matters (Ma, 2005). In this sense, the administrative restructuring process in the post-reform period is actually a ‘reterritorialisation’ of the state, with the rescaling of the state power from the predominant national level to different levels of local governments (Wu, 2002). Owing to the importance of the administrative level, Chinese cities have the strongest incentive to upgrade their administrative levels. The State Council is empowered by the Chinese constitution to control urban administrative level upgrading, whilst the Ministry of Civil Affairs acts as the de facto regulator of urban administrative level upgrading (Shen, 2007).

In terms of administrative levels, Chinese cities can be classified into four categories, namely, (1) province-level cities, (2) vice-province-level cities, which are a half-level lower than province-level units but a half-level higher than prefecture-level units, (3) prefecture-level cities, and (4) county-level cities. By implementing an urban administrative level upgrading strategy, many cities and counties were upgraded to high administrative levels. Upgrading cities to the top administrative levels (i.e. provincial and vice-provincial level) reflects the national development strategy of China, although this rarely happens. Chongqing is the only city that was upgraded to the provincial level in 1997. In addition, 16 cities were upgraded to the vice-provincial level in 1994. These cities were granted the privilege of listing their economic plans separately from those of their provinces to allow them to break free from the rigid administrative hierarchy (Solinger, 1991).¹ At low administrative levels, on the one hand, many prefecture-level cities were established to lead the urban–rural integration and economic development of their surrounding regions.² On the other hand, many counties with urban characteristics were upgraded to county-level cities.³ As very few cities have been upgraded to the vice-provincial level and above, we focus on the county- to prefecture-level city upgrading and county to county-level city upgrading. Table 1 shows the changes in the number of China’s administrative units at different levels from 1978 to 2015. As can be seen, the number of prefecture-level cities increased from 99 in 1978 to 276 in 2015, whereas the number of county-level cities increased from 91 to 361 during the same period.

Table 1.

Changes in the number of China’s administrative units at three levels from 1978 to 2015

	1978	1990	2000	2010	2015
Provincial level
Provinces	21	22	22	22	22
Autonomous regions (zizhi qu)	5	5	5	5	5
Centrally administered cities	3	3	4	4	4
Total	29	30	31	31	31

Prefecture level
Vice-provincial-level cities	0	0	15	15	15
Prefecture-level cities	99	185	244	268	276
Prefectures (diqu)	206	113	37	17	10
Autonomous prefectures (zizhi zhou)	28	30	30	30	30
Leagues (meng)	7	8	7	3	3
Total	340	336	333	333	334

County level
County-level cities	91	279	400	370	361
Counties	1990	1731	1503	1461	1397
Urban districts (shiqu)	408	651	787	853	921
Autonomous counties (zizhi xian)	65	121	116	117	117
Banners (qi)	83	51	49	49	49
Total	2637	2833	2855	2850	2845

Source: Data of 2000, 2010 and 2015 are summarised by the authors based on figures from the Ministry of Civil Affairs of China (available at: http://xzqh.mca.gov.cn/statistics/, accessed 14 January 2019); data of 1978 and 1990 are derived from NBS (2010); the number for Hong Kong, Macau and Taiwan is not included.

The establishment of prefecture-level cities has been carried out mostly in three ways, namely, merging prefecture-level cities with prefectures (dishi hebing), upgrading county-level cities to prefecture-level cities and upgrading counties to prefecture-level cities (for details, see Ma, 2005). The first way does not change the administrative level of cities but extends their administrative geographical areas, whilst the third way is rarely implemented in practice. Therefore, we focus on the second way, that is, county- to prefecture-level city upgrading.

County- to prefecture-level city upgrading is generally implemented by abolishing a prefecture and simultaneously promoting a county-level city located in the prefecture to the rank of prefecture-level city. Several major territorial rearrangements have been carried out in this type of administrative restructuring. First, the upgraded county-level city, usually the economic centre of the prefecture, is converted to the city-administered district(s) (shixiaqu), which is the central city (zhongxin chengqu) of the new prefecture-level city.⁴ Second, the rest of the prefecture’s counties or county-level cities are placed under the administrative ‘control’ (guan) of the newly created prefecture-level city, which is known as the ‘city administering county’ (shi guan xian) system. Third, the government offices of the former prefecture and the upgraded county-level city are reorganised, after which new departments and bureaus on public affairs, urban construction, industrial and commercial administration and housing and real estate are established in order to enhance the urban managing functions of the new prefecture-level city (Chung, 2007). This upgrading strategy has been widely implemented in many areas since 1978 and has largely increased the number of prefecture-level cities.

We use the case of Dezhou City in Shandong Province to elaborate the territorial rearrangement in the county- to prefecture-level city upgrading. The county-level city of Dezhou was upgraded to prefecture-level in 1994 and Dezhou Prefecture was abolished at the same time. Before upgrading, Dezhou Prefecture comprised two county-level cities (Dezhou and Leling) and nine counties (Figure 1). After upgrading, the territory of the former county-level Dezhou city was converted to Decheng district, which is the central city of the prefecture-level city of Dezhou. The other ten county-level units were placed under this newly established city.

Figure 1.

County- to prefecture-level city upgrading: Dezhou’s case

The upgrading of a county to a county-level city is commonly implemented by converting the whole county to county-level city (xian gai shi or che xian gai shi). County to county-level city upgrading does not enlarge the territory of the county, and the entire county is labelled a ‘city’ after upgrading. Although upgrading to county-level city does not actually change a county’s rank in the administrative system, the upgrading can reclassify the county into an urban-based territorial unit and the government gains more political power. County to county-level city upgrading has been the dominant way to establish new cities since the early 1980s.

An official rule was implemented to regulate county- to prefecture-level city and county to city upgrading. The official rule for the establishment of a county-level city appeared in 1983. Owing to low minimum requirements, over 100 counties obtained city status from 1983 to 1986. The central government raised the minimum requirements in 1986 and 1993 because the numbers of county to county-level city upgrading continued to rise. In 1993, more systematic requirements for the establishment of county- and prefecture-level cities were proposed by the Ministry of Civil Affairs (Table 2). The newly issued requirements set different standards for counties with different population densities as well as the minimum socioeconomic requirements for establishing prefecture-level cities. However, the criteria are not rigidly applied in practice. Under the 1993 requirements, about 100 counties obtained city status from 1994 to 1997, as the number of prefecture-level cities had increased steadily since the early 1980s. After 1997, the central government suspended the massive upgrading of counties to county-level cities, because the newly upgraded cities resulted in a massive fiscal burden. Therefore, the number of county-level cities slowly increased in the late 1990s and declined after 2000, during which time county-level cities were converted into urban districts. According to Li (2011), the upgrading process is not automatic for counties/cities that satisfy the minimum requirements, but is taken as a policy instrument to give local governments the right incentive to promote economic development.

Table 2.

Minimum requirements for city designation issued in 1993

		Population density (person/km²)
		< 100	100–400	> 400
Minimum requirements for the establishment of county-level city
Urbanisation level	Non-agricultural population (person)	100,000	120,000	150,000
Urbanisation level	Share of non-agricultural population (%)	20	25	30
Industrialisation level	Industrial output value of towns and above (million yuan)	800	1200	1500
	Share of industrial output value in gross output value of industry and agriculture (%)	60	70	80
	Share of output value of tertiary sector in gross domestic output (%)	20	20	20
	Gross domestic output (million yuan)	600	800	1000
Fiscal strength	Revenue in the general public budget obtained at the local level (million yuan)	40	50	60
	Per capita fiscal revenue (yuan)	60	80	100
	To transfer amount of fiscal revenue to the governments at higher levels
Minimum requirements for the establishment of prefecture-level city
Non-agricultural population of the urban districts (person)		250,000
Non-agricultural population of the urban districts where the city government is located (person)		200,000
Gross output value of industry and agriculture (million yuan)		3000
Share of industrial output value in gross output value of industry and agriculture (%)		80
Gross domestic output (million yuan)		2500
Share of output value of tertiary sector in gross domestic output (%)		35 (should be larger than the ratio of the first sector)
Revenue in the general public budget obtained at the local level (million yuan)		200

Source: Summarised by the authors based on The Report on Adjusting the Criteria for the Designation of New Cities in 1993 proposed by the Ministry of Civil Affairs.

Reorganisation of power relations in urban administrative level upgrading

Urban administrative level upgrading strategy has reorganised the power relations of territorial units, in a process involving a shift in administrative and economic powers from surrounding low-level territorial units to cities. Placing cities at higher levels allows the cities with governments that have more abilities to obtain and mobilise various resources related to urban development. After upgrading to a high administrative level, cities enjoy greatly expanded powers to use various administrative tools, such as land, investments and industrial policies, in order to develop their economies. In Table 3, we provide a systematic summary of the upward shift of state power in county- to prefecture-level city and county to county-level city upgrading. Although the changes are quite intricate and vary between provinces, upgrading can enhance the power of the upgraded cities in three major aspects, which are administrative powers, fiscal benefits and policy priority (see Table 3 for details).

Table 3.

Upward shift in the state power of urban administrative level upgrading^a

Category	Upgrading type	Upwards shift in state power	Source
Administrative powers	County- to prefecture-level city upgrading	•Owing to the ‘city administering county’ system, prefecture-level cities are the intermediate administration between provinces and counties, enjoying expanded powers to administer several county-level cities, counties and their equivalents within their jurisdiction•Prefecture-level cities have the decision-making power in administrative, fiscal and economic matters, while subordinate administrative units, counties or county-level cities have to follow the interests of their superior prefecture-level cities•Prefecture-level cities have great privilege to control the resources (e.g. investments, development projects and land development quotas) of subordinate county-level cities and counties	Chung (2007); Dai (2000); Ma (2005)
	County to county-level city upgrading	•County-level cities have more authority on foreign trade, exchange management, police recruitment and could approve projects with a higher cap of investment	Chung and Lam (2004); Dai (2000)
Fiscal benefits	County- to prefecture-level city upgrading	•Prefecture-level cities can enjoy a wider range of taxes and fees than county-level cities•Prefecture-level cities have the power of fiscal resources allocation and to maximise and share a part of fiscal extractions from the counties and county-level cities under their jurisdiction	Chung and Lam (2004); Wong and Bird (2008); Zhou (2012)
	County to county-level city upgrading	•County-level cities enjoy powers to collect higher urban construction tax ratios (7%) compared with counties (5%)•County-level cities are likely to receive more quotas to convert farm land to urban construction land, through which city governments can gain a large amount of fiscal revenue	Chung and Lam (2004); Lichtenberg and Ding (2009)
Policy priority	County- to prefecture-level city upgrading	•Prefecture-level cities are more likely to acquire preferential policies from the central government and provinces in which they are located^b•The policy priority extends policy-making powers of prefecture-level cities to develop their own economies•Prefecture-level cities have greater bargaining power than county-level cities and counties when competing for foreign and domestic investments	Ma (2005); Zhang and Zhao (1998)
	County to county-level city upgrading	•County-level cities will be listed separately in the provincial plan and thereby are more likely to seek preferential policies and investment projects from the superior governments than counties•Compared with counties, county-level cities gain greater prestige and are more attractive to investors	Chung and Lam (2004); Shen (2007); Zhang and Zhao (1998)

Notes: ^a This is an incomplete list summarised on the basis of existing literature.

Preferential policies typically include different types of development zones at the national level, such as Economic and Technological Development Zones, High-Tech Industrial Development Zones, Bonded Areas, Export Processing Zones, Coastal Open Economic Zones and other special development zones.

First, county- to prefecture-level city upgrading has fundamentally expanded the administrative powers of cities because prefecture-level cities serve as the intermediate administration between provinces and subordinate county-level cities or counties. Prefecture-level cities can take advantage of their administrative powers to play a powerful role in basic decision-making related to resource allocation, key industrial, transportation and energy projects as well as general directions and development plans (Chung, 2007; Dai, 2000; Ma, 2005). Second, the fiscal power of prefecture-level cities has been strengthened after upgrading because they enjoy a wide range of taxes and fees and play a dominant role in fiscal resource allocation (Chung and Lam, 2004; Wong and Bird, 2008; Zhou, 2012).⁵ Third, county- to prefecture-level city upgrading provides the newly upgraded prefecture-level cities with better policy priority and a broader range of policy-making powers. The policy priority has enabled prefecture-level cities to carry out their development strategies (Ma, 2005; Zhang and Zhao, 1998).

By contrast, county to county-level city upgrading expands the state power of counties to a lesser extent (Table 3). For example, county-level cities have more administrative powers to govern urban-related affairs (e.g. foreign trade, exchange management, police recruitment, vehicle administration and investment projects approval) (Chung and Lam, 2004; Dai, 2000). In terms of fiscal benefits, county-level cities have the power to collect extra taxes and fees related to urban construction and management (Zhang and Zhao, 1998). Compared with counties, county-level cities are more likely to secure more quotas from the superior government to convert farm lands into construction lands, which can increase fiscal revenue (Lichtenberg and Ding, 2009). In addition, county-level cities are more likely to have policy priority than counties, as they will be listed separately in provincial plans (Shen, 2007). After obtaining city status, county-level cities enjoy higher prestige compared with counties, thereby allowing the former to attract outside investors (Chung and Lam, 2004). However, these benefits may not lead to significant differences in the state power of county-level cities and counties. In other words, obtaining city status may not provide the county-level cities with endogenous driving forces, because the status does not actually change a county’s rank in China’s administrative division system. In this sense, county to county-level city upgrading has less to gain from the upward shift in state power than county- to prefecture-level city upgrading.

We argue that the upgraded cities are granted great state power in administrative, fiscal, policy-making, production and other matters, thereby allowing these cities to obtain and mobilise various resources related to development. Given the weakness of the private sector and the absence of a strong civic society in China, the state power of the urban government is critical in achieving urban development in the post-reform period. Although urban governments no longer directly intervene in the operation of local enterprises, they can affect economic development by using policy tools, such as land, taxes and permits. Cities at high administrative levels are more likely to attract domestic and foreign investments, acquire land for construction, increase fiscal revenue and initiate new development projects that can promote development. Therefore, urban administrative level upgrading is believed to have a positive effect on the development performance of Chinese cities.

Empirical analysis of the effects of urban administrative level upgrading on urban development performance

Hypotheses

Our empirical analysis leads to two testable hypotheses. First, the state power is important for urban development in China. Urban administrative level upgrading has expanded the state power of these upgraded cities. Therefore, the following hypothesis is proposed:

Hypothesis 1: Urban administrative level upgrading has a positive effect on the development performance of cities.

Second, China’s administrative division system continues to be hierarchical and level-based in the post-reform period. A close correspondence exists between the administrative level of a city and its ability to promote its development. Although county- to prefecture-level city upgrading and county to county-level city upgrading involve the upward shift of state power, the former changes the administrative level of upgraded cities whilst the latter does not. Therefore, we propose the following hypothesis:

Hypothesis 2: The two types of upgrading, namely, county- to prefecture-level city upgrading and county to county-level city upgrading, have different effects on the development performance of upgraded cities.

Methods

We apply a quasi-experimental method that combines PSM with the DID approach to estimate the effects of the upgrading on urban development performance. The key challenge in comparing the performance of upgraded cities and other cities/counties is the endogeneity problem in the choice of upgraded cities. Direct comparisons of the performance of upgraded cities and other cities/counties are problematic because cities/counties with better development performance may inherently have greater opportunities to be upgraded. This situation makes it difficult to estimate the net effect of upgrading. PSM is used widely to construct control groups for estimating causal effects in non-experimental settings, which can correct for biases in sample selection, such as those caused by self-selection or systematic judgment of the researcher (Imbens and Wooldridge, 2009; Rosenbaum and Rubin, 1983). Following the ‘selection on observables’ assumption, PSM pairs treatment and control units based on similar measures that are calculated from various observable characteristics (Abadie and Imbens, 2006). The nature of propensity score is ‘the conditional probability of assignment to a particular treatment given a vector of observed covariates’ (Rosenbaum and Rubin, 1983: 41).

In this study, urban administrative level upgrading is considered as a ‘treatment’ that does not affect all cities in the same way. The covariates included in calculating the propensity score should capture all factors affecting the upgrading decision. Thus, we chose covariates based on the minimum requirements for city designation (Table 2) to control for known factors for urban administrative level upgrading. With the propensity score calculated from logistic regression, we matched each county/city where the administrative level has been upgraded ex post (treatment) to a county/city that was not upgraded ex post, but has an approximately equal likelihood (i.e. propensity score) to be upgraded ex ante (control) using STATA’s prmatch2 (Leuven and Sianesi, 2003).

After constructing the treatment and control groups using PSM, we employed the DID approach and tracked changes in urban development performance for four years after upgrading. Ideally, we are able to compare the performance of the same city/county with upgrading and without; however, we cannot observe the counterfactual performance of the upgraded cities/counties if they have not upgraded (or the converse). Thus, we used the DID approach to assess the causal effect by comparing the average change in the urban development performance of the treatment and control groups before and after upgrading (Angrist and Pischke, 2008). In performing such a step, we are able to estimate the net effect by taking the development performance of the control group as the counterfactual performance of the treatment group. Following the approach of Chang et al. (2013a, 2013b), we used the average treatment effect on the treated (ATT) to measure the performance difference between the two groups of cumulative changes in urban development performance. ATT is calculated by equation (1):

AT T_{k} = \frac{1}{n} \sum^{(Y_{t + k}^{treated} - Y_{t + k}^{control})} - \frac{1}{n} \sum^{(Y_{t}^{treated} - Y_{t}^{control})}

(1)

where Y denotes the variable measuring urban development performance; k represents one, two, three or four years since upgrading; and n is the number of matches. The corresponding standard errors are calculated using equation (2).

S E_{k} = \sqrt{\frac{1}{n} Var (Y_{t + k} - Y_{t} | treatment = 1) + \frac{1}{n} Var (Y_{t + k} - Y_{t} | control = 1)}

(2)

A simple t-test is used to examine whether the cumulative change between the treatment and control groups in urban development performance is significant.

Data and variables

Figure 2 presents the number of upgrading cases between 1984 and 2010. County- to prefecture-level city upgrading shows a roughly steady trend during the entire reform era. Two major waves of county to county-level city upgrading are observed in 1987–1988 and 1994–1997. Given that statistical data on county-level units for the period before 1990 are almost unavailable, we focused on the urban administrative level upgrading during the period since 1994. The upgrading in this period is regulated by the 1993 requirements. The data used in this study were collected from the yearly series of Sub-Provincial Public Finance Statistics published by the Ministry of Finance. The data covered all county-level territorial units and included detailed public finance and basic socio-economic information, and the first year of the published data was 1993.

Figure 2.

Number of urban administrative level upgrading cases from 1984 to 2010

We restricted our sample to cities for which urban administrative upgrading did not involve geographical expansion and annexation during the research period. This was done to exclude the effects related to statistical reasons. According to this constraint, 82 county to county-level city upgrading cases were identified during the selected treatment period (1994 to 1997). In terms of county- to prefecture-level city upgrading, we set the treatment period as 1994 to 1998, in which 23 upgrading cases are in our sample. These two samples of cities represent the treatment groups for each type of urban administrative level upgrading. Two control groups are matched correspondingly from all county-level cities and counties that remained intact from 1993 to 2002. In addition, we set the relative time (t−1, t, t+1, t+2, t+3 and t+4) for each pair of cities/counties to denote the chronological distance from the treatment start year. We measured changes in urban development performance from year t−1 through year t+4.

We used three variables to measure the development performance of cities, namely, urban non-agricultural population, industrial output value and budgetary fiscal revenue. It should be noted that we employed the concept of urban development in a broad sense, because the upgrading can influence various aspects of the development of cities. During the 1990s, the urban non-agricultural population could be used as a proxy of urban population because the amount of floating population was not large. As statistics on GDP are not available at the county level, industrial output value is a better measure of city-oriented development performance than the gross value of industrial and agricultural outputs. A widely held view among Chinese government leaders in the 1990s is that industries play a key role in transforming a city/county into a wealthy one (Ma, 2005). Fiscal revenue is thus used to capture the increase in fiscal strength of the upgraded cities/counties.

To clarify the causal relationships, we included the lagged (1 year prior to upgrading) urban characteristics variables based on the 1993 requirements when calculating the propensity score. The basic assumption is that counties/cities are more likely to receive an upgrade if they meet those requirements. In terms of county to county-level city upgrading, the three major requirements are industrialisation level, urbanisation level and fiscal strength. We used nine variables to capture the three dimensions of the requirement: gross value of industrial and agricultural output (GVIAO), fiscal revenue (FISREV), per capita fiscal revenue (PCFISREV), total population (TOTALPOP), urban non-agricultural population (URBANPOP), urbanisation ratio (URBANIZATION), industrial output value (INDOUTPUT), share of industrial output in gross output (INDSHARE) and transferred fiscal revenue to the governments at higher levels (FISTRANSFER). For county- to prefecture-level city upgrading, the official requirements are less than those of county to county-level city upgrading. Seven variables are included in logistic regression, namely, GVIAO, FISREV, TOTALPOP, URBANPOP, INDSHARE, INDOUTPUT and share of tertiary employment in total urban employment (TERRITIARYSHARE). All the characteristics are measured at the same scale before and after upgrading. In county- to prefecture-level city upgrading, the upgraded county-level city is converted into the urban district(s) of the new prefecture-level city; thus, the variables of the post-upgrading period are measured by the characteristics of urban district(s). Hence, these variables did not include subordinate county-level cities and counties of newly established prefecture-level cities.

Results

Table 4 displays the logistic regression results for the two types of urban a administrative level upgrading using a pair matching (one-to-one matching) method. We include year fixed and region effects in matching the control group for county to county-level city upgrading to control for unobservable factors and subsequently improve the matching precision.⁶ However, we only include year fixed effects when matching the control group for county- to prefecture-level city upgrading because of the small reservoir of potential controls. We eventually generated 23 and 82 pairs of cities/counties for county- to prefecture-level city upgrading and for county to county-level city upgrading, respectively. Table 4 further shows the balancing tests for the matching results. The results from the pairing t-test and Hotelling test confirm that our matching is well balanced, thereby confirming our hypothesis that the cities/counties of the treatment and control groups have approximately the same likelihood of being upgraded in the year prior to upgrading.

Table 4.

Balancing tests for the treatment and control groups of the two types of urban administrative level upgrading

Variables	Logistic estimation for propensity scores	t-test on the mean of each matching variable
Variables	Logistic estimation for propensity scores	Treatment	Control	t-stat	p-value
County- to prefecture-level city upgrading
GVIAO (t−1)	−1.26e2212−06(8.83e−07)	7.0e+05	6.3e+05	0.37	0.713
FISREV (t−1)	0.0000831(0.000281)**	17,377	13,830	0.96	0.340
TOTALPOP (t−1)	0.00613(0.0071)	78.988	71.064	0.59	0.557
INDSHARE (t−1)	−0.34(0.553)	0.78	0.75	0.58	0.566
URBANPOP (t−1)	0.094(0.0265)***	22.19	21.15	0.38	0.705
TERRITIARYSHARE (t−1)	0.0315(0.0212)	27.86	32.07	−1.07	0.292
INDOUTPUT (t−1)	−4.61e−09(8.09e−07)	5.7e+05	6.2e+05	−0.28	0.784
Region fixed effects	Not included
Year fixed effects	Included
Intercept	−7.106(1.009)***
N	1343	23	23
Pseudo R²	0.185	Hotelling test F = 0.543 (p-value 0.796)
County to county-level city upgrading
GVIAO (t−1)	1.11e−06(1.63e−06)	3.9e+05	4.1e+05	−0.39	0.694
FISREV (t−1)	−3.63e−06(3.34e−05)	8400.8	8991.0	−0.55	0.585
PCFISREV (t−1)	0.00334(0.00185)	151.74	152.13	−0.02	0.984
TOTALPOP (t−1)	0.00985(0.0062)	64.43	66.35	−0.36	0.720
URBANPOP (t−1)	0.106(0.0326)***	10.02	9.83	0.22	0.824
URBANIZATION (t−1)	2.945(1.18)**	0.20	0.18	0.58	0.561
INDOUTPUT (t−1)	−5.28e−07(2.00e−06)	3.0e+05	3.4e+05	−0.51	0.613
INDSHARE (t−1)	4.267(0.882)***	0.70	0.73	−1.00	0.317
FISTRANSFER (t−1)	3.43e−5(3.66e−5)	6338.1	6328.8	0.01	0.992
Region fixed effects	Included
Year fixed effects	Included
Intercept	−8.413(0.770)***
N	4858	82	82
Pseudo R²	0.263	Hotelling test F = 0.356 (p-value 0.953)

Note: ** and *** indicate significance at p < 0.01 and p < 0.001, respectively.

Table 5 compares the urban development performance of 23 upgraded cities with that of the matched cities. In general, upgraded cities in the treatment group experienced faster development compared with the remaining county-level cities in the control group during the post-upgrading period. For the urban non-agricultural population, upgraded cities experience an average increase that has been greater than that of remaining county-level cities since the year of upgrading. ATT increased to 1.84 and 3.01 three and four years after upgrading, respectively, and these estimates are all significant at the 1% level. The result demonstrates that county- to prefecture-level city upgrading positively affects urban population growth and the effect becomes significant three years after upgrading. Moreover, we find similar results when comparing budgetary fiscal revenue. ATTs are positive and significant two years after upgrading, showing that county- to prefecture-level city upgrading positively increases the fiscal strength of the newly transformed prefecture-level city. By contrast, ATTs are negative and insignificant regarding the industrial output value. ATTs decrease from 53,255.2 to −137,341.9 from the year of upgrading to four years after upgrading. The average increase in the industrial output of the control group cities is larger than that of the upgraded cities although the difference is not significant in statistics. This result suggests that county- to prefecture-level city upgrading could not promote the industrialisation of the upgraded cities. This result might be attributed to the effect of the conversion of county-level cities to the urban districts, whereby the districts are likely to function as service and administration centres instead of industrial bases.

Table 5.

Urban development performance of the treatment and control groups for the two types of urban administrative level upgrading

Urban non-agricultural population
Year	t−1	t	t+1	t+2	t+3	t+4
County- to prefecture-level city upgrading
Treatment	22.19	22.90	23.71	24.52	25.79	27.08
Control	21.15	21.58	22.08	22.55	22.91	23.03
ATT		0.28	0.59	0.93	1.84**	3.01**
S.E.		0.19	0.39	0.58	0.70	0.88
# Matches	23	23	23	23	23	23
County to county-level city upgrading
Treatment	10.02	10.55	10.84	11.18	11.66	12.00
Control	9.83	10.12	10.46	10.80	11.38	11.56
ATT		0.24	0.19	0.19	0.09	0.25
S.E.		0.37	0.32	0.38	0.41	0.46
# Matches	82	82	82	82	82	82
Urban industrial output value
County- to prefecture-level city upgrading
Treatment	565,494.4	589,891.0	601,940.7	611,151.2	625,552.4	682,191.7
Control	618,278.3	589,419.7	674,529.6	747,695.6	793,685.0	872,317.5
ATT		53,255.2	−19,805.0	−83,760.5	−115,348.7	−137,341.9
s.e.		114,994.8	131,813.6	149,508.1	160,503.9	180,495.3
# Matches	23	23	23	23	23	23
County to county-level city upgrading
Treatment	303,091.7	398,217.4	487,105.3	491,665.7	524,425.6	579,700.8
Control	336,657.3	394,426.5	469,254.1	493,671.5	552,618.0	566,241.2
ATT		37,356.5	51,416.8	31,559.8	5373.2	47,025.2
s.e.		30,464.2	40,509.8	48,499.7	60,372.2	79,668.7
# Matches	82	82	82	82	82	82
Urban budgetary fiscal revenue
County- to prefecture-level city upgrading
Treatment	17,377.2	22,609.4	26,668.1	29,880.4	33,661.7	37,603.1
Control	13,830.2	13,879.7	16,072.9	18,385.0	20,858.1	22,934.0
ATT		5182.7	7048.2	7948.4*	9256.6*	11,122.1*
s.e.		3402.1	3597.8	3854.1	4533.6	5479.9
# Matches	23	23	23	23	23	23
County to county-level city upgrading
Treatment	8400.8	7552.8	9236.6	10,686.4	11,351.4	12,845.2
Control	8991.0	7704.8	9036.3	10,974.2	11,760.6	13,348.9
ATT		438.2	790.5	302.4	181.0	86.5
s.e.		746.2	733.3	792.0	845.7	874.9
# Matches	82	82	82	82	82	82

Note: * and ** indicate significance at p < 0.05 and p < 0.01, respectively.

By contrast, the results demonstrate that county to county-level city upgrading cannot improve the urban development performance at least in the short term. For all three variables measuring urban development performance, ATTs are positive and insignificant. The results indicate that the average increases in the urban non-agricultural populations, industrial outputs and fiscal revenues of newly created county-level cities are not significantly higher than those of the remaining counties in the control group. In other words, the newly created county-level cities experience an average increase in the urban development performance, which is slightly greater than the remaining counties from one year prior to upgrading through four years after upgrading. However, the upgrading may not lead to the performance-enhancing effects on the upgraded cities in the short term.

The results suggest that the two types of urban administrative level upgrading have different effects on urban development performance. The upgrading to prefecture-level cities might lead to significant increases in urban population and fiscal revenue compared with their remaining intact counterparts. However, the upgrading does not affect the industrialisation of the newly established prefecture-level cities. We argue that county- to prefecture-level city upgrading has immediate performance-enhancing effects on these newly upgraded cities in the short term, whereas county to county-level city might not have such effects at least in the short term. For county to county-level city upgrading, cities/counties in the treatment and control groups exhibit roughly parallel development trends at around four years after upgrading.

The difference between the effects of these two upgrading types on urban development performance can be attributed to the extent to which upgrading has changed the state power of the cities. County- to prefecture-level city upgrading changes the administrative level of upgraded cities, essentially expanding their state power. After upgrading, these cities serve as regional centres and benefit from their central role within their jurisdictions. Therefore, county- to prefecture-level city upgrading produces immediate effects on the development of newly upgraded cities. By contrast, county to county-level city upgrading does not lead to fundamental changes in the state power of newly upgraded cities, because counties and county-level cities are at the same administrative level and are governed administratively by their superior prefecture-level units. Therefore, county to county-level city upgrading has no immediate effects on urban development performance. As a result, the upgrading to county-level cities cannot improve their driven forces for urban development.

Discussion and conclusion

The present study contributes to the literature on the administration/economy interface through an empirical investigation of the processes involved in the implementation of urban administrative level upgrading in post-reform China. As a distinctive institution originally designed for socialist territorial administration, China’s administrative division system has retained its rigid hierarchical and level-based structure in the post-reform period. Since the early 1980s, China has decided to make cities the engine of economic growth. Urban administrative level upgrading has been implemented as an important administrative restructuring strategy to place selected cities at high administrative levels within the administrative division system and expand the state power of these cities. Given the deep involvement of urban governments in China’s urbanisation and urban development, the state power of cities determines the ability of urban governments to effectively affect urban development performance.

This study empirically examines whether urban administrative level upgrading leads to better urban development performance. Three variables, namely, urban non-agricultural population, industrial output value and budgetary fiscal revenue, are used to measure urban development performance. The results indicate that county- to prefecture-level city upgrading positively and significantly increases urban population growth and budgetary fiscal revenue in three and two years after upgrading, respectively, but these do not necessarily lead to rapid industrialisation. However, this finding is not true for county to county-level city upgrading. Upgraded counties might not outperform remaining counties in terms of urban non-agricultural population, industrial output and budgetary fiscal revenue. County- to prefecture-level city upgrading has empowered the newly founded prefecture-level cities to serve as regional centres, essentially expanding their state power. In turn, the expansion of state power as a result of the upgrading significantly affects the development of newly established regional central cities. By contrast, county to county-level city upgrading does not lead to fundamental expansion of state power. These newly upgraded county-level cities experience roughly parallel development with remaining counties in the control group. Such a difference may be attributed to their being at the same administrative level as the county and county-level cities, whereas prefecture-level cities are at a higher administrative level compared with county-level cities and counties. These results, in turn, confirm that China’s administrative division system, although restructured, remains hierarchical and level-based in the post-reform era, and that urban administrative level upgrading can lead to better urban development performance.

Western developed countries may be compelled to rescale state power to deal with the challenges of neoliberal globalisation, whilst China appears to restructure the administrative division system to accomplish geographically or spatially targeted urbanisation and urban development. With a level-based administrative division system, the Chinese central government can effectively influence the development of cities by reorganising the distribution of state power among cities at different administrative levels. The continuing strategic role of the state in China’s state-led urbanisation model has been emphasised with the establishment of new scales and arenas, such as the newly upgraded prefecture-level cities, and with the adjustment of the administrative structures of the state. For example, urban administrative level upgrading is used as an important policy instrument to implement city-centred administrative restructuring, thereby reflecting the omnipresent regulatory intervention of the Chinese state in urbanisation and urban development. We argue that the hierarchical and vertically linked nature of the administrative division system, which is absent in Western developed countries, enables the Chinese state to dominate China’s urbanisation process in the post-reform period.

This study has several limitations. For example, the data set provides limited information on urban development. Thus, we are unable to examine the long-term effects of the upgrading on the development performance in more aspects. Moreover, owing to the complex mechanisms of urban administrative level upgrading, detailed investigations, such as those involving more case studies, are needed to provide in-depth knowledge. Finally, county to county-level city upgrading seems to have been re-initiated in recent years. Thus, further research is needed in order to fully investigate the effects of administrative level upgrading on development of Chinese cities at the new stage of urbanisation.

Footnotes

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 work was supported by the National Science Foundation of China (No. 41701182); and Chan To-Haan Endowed Professorship Fund and Distinguished Research Achievement Award of the University of Hong Kong; and GRF (No. HKU-17209014) of the Research Grants Council of Hong Kong.

ORCID iDs

Jiejing Wang

Anthony GO Yeh

Notes

References

Abadie

Imbens

(2006) Large sample properties of matching estimators for average treatment effects. Econometrica 74(1): 235–267.

Angrist

Pischke

J-S

(2008) Mostly Harmless Econometrics: An Empiricist’s Companion. Princeton, NJ, and Oxford: Princeton University Press.

Bairoch

(1988) Cities and Economic Development: From the Dawn of History to the Present. Chicago, IL: University of Chicago Press.

Bennett

(1997) Administrative systems and economic spaces. Regional Studies 31(3): 323–336.

Berry

BJL

(1981) Comparative Urbanization: Divergent Paths in the Twentieth Century. New York: St. Martin’s Press.

Brenner

(1999) Globalisation as reterritorialisation: The re-scaling of urban governance in the European Union. Urban Studies 36(3): 431–451.

Cartier

(2013) What’s territorial about China? From geopolitical narratives to the ‘administrative area economy’. Eurasian Geography and Economics 54(1): 57–77.

Cartier

(2015) Territorial urbanization and the party-state in China. Territory, Politics, Governance 3(3): 294–320.

Chan

(2010) Fundamentals of China’s urbanization and policy. The China Review 10(1): 63–93.

10.

Chang

Chung

Moon

(2013a) When do foreign subsidiaries outperform local firms? Journal of International Business Studies 44(8): 853–860.

11.

Chang

Chung

Moon

(2013b) When do wholly owned subsidiaries perform better than joint ventures? Strategic Management Journal 34(3): 317–337.

12.

Chien

(2013) New local state power through administrative restructuring – A case study of post-Mao China county-level urban entrepreneurialism in Kunshan. Geoforum 46: 103–112.

13.

Chung

(2007) The change in China’s state governance and its effects upon urban scale. Environment and Planning A 39(4): 789–809.

14.

Chung

Lam

(2004) China’s ‘city system’ in flux: Explaining post-Mao administrative changes. The China Quarterly 180: 945–964.

15.

Dai

(2000) City System of China. Beijing: China Map Press (in Chinese).

16.

Fitzgerald

(2003) The province in history. In: Fitzgerald

(ed.) Rethinking China’s Provinces. New York: Routledge, pp. 12–39.

17.

Harvey

(2005) A Brief History of Neoliberalism. Oxford: Oxford University Press.

18.

(2009) China’s emerging neoliberal urbanism: Perspectives from urban redevelopment. Antipode 41(2): 282–304.

19.

Imbens

Wooldridge

(2009) Recent developments in the econometrics of program evaluation. Journal of Economic Literature 47(1): 5–86.

20.

Jessop

(2002) Liberalism, neo-liberalism and urban governance: A state theoretical perspective. Antipode 34(3): 452–472.

21.

Jessop

(2004) Hollowing out the ‘nation-state’ and multilevel governance. In: Kennett

(ed.) A Handbook of Comparative Social Policy. Cheltenham: Edward Elgar, pp. 11–25.

22.

Kirkby

RJR

(1985) Urbanisation in China: Town and Country in a Developing Economy, 1949–2000 AD. London: Croom Helm.

23.

Leuven

Sianesi

(2003) PSMATCH2: Stata module to perform full Mahalanobis and propensity score matching, common support graphing, and covariate imbalance testing. Available at: http://ideas.repec.org/c/boc/bocode/s432001.html (accessed 25 March 2019).

24.

(2011) The incentive role of creating ‘cities’ in China. China Economic Review 22(1): 172–181.

25.

(2015) State rescaling and national new area development in China: The case of Chongqing Liangjiang. Habitat International 50: 80–89.

26.

(2012) The transformation of regional governance in China: The rescaling of statehood. Progress in Planning 78(2): 55–99.

27.

Yeh

AGO

(2014) State rescaling and the making of city-regions in the Pearl River Delta, China. Environment and Planning C: Government and Policy 32(1): 129–143.

28.

Lichtenberg

Ding

(2009) Local officials as land developers: Urban spatial expansion in China. Journal of Urban Economics 66(1): 57–64.

29.

Liu

(1996) Theory and Practice in Administrative Zoning in China. Shanghai: East China Normal University Press (in Chinese).

30.

Liu

Yin

(2012) Local state and administrative urbanization in post-reform China: A case study of Hebi City, Henan Province. Cities 29(2): 107–117.

31.

Luo

Cheng

Yin

, et al. (2014) Province-leading-county as a scaling-up strategy in China: The case of Jiangsu. The China Review 14(1): 125–146.

32.

LJC

(2002) Urban transformation in China, 1949–2000: A review and research agenda. Environment and Planning A 34(9): 1545–1569.

33.

LJC

(2005) Urban administrative restructuring, changing scale relations and local economic development in China. Political Geography 24(4): 477–497.

34.

McGee

(1998) Governing mega-urban regions: The case of Vancouver. In: Friedmann

(ed.) Urban and Regional Governance in the Pacific. Vancouver: University of British Columbia Press, pp. 113–115.

35.

McGinnis

(1999) Polycentric Governance and Development. Ann Arbor, MI: University of Michigan Press.

36.

Maurel

M-C

(1989) Administrative reforms in Eastern Europe: An overview. In: Bennett

(ed.) Territory and Administration in Europe. London: Frances Pinter, pp. 111–123.

37.

National Bureau of Statistics (NBS) (2010) China Compendium of Statistics, 1949–2008. Beijing: China Statistical Press (in Chinese).

38.

O’Brien

(1992) Global Financial Integration: The End of Geography. London: Pinter.

39.

(1993) Reform and urban bias in China. The Journal of Development Studies 29(4): 129–148.

40.

Ong

(2014) State-led urbanization in China: Skyscrapers, land revenue and ‘concentrated villages’. The China Quarterly 217: 162–179.

41.

Rosenbaum

Rubin

(1983) The central role of the propensity score in observational studies for causal effects. Biometrika 70(1): 41–55.

42.

Shen

(2006) Understanding dual-track urbanisation in post-reform China: Conceptual framework and empirical analysis. Population Space and Place 12(6): 497–516.

43.

Shen

(2007) Scale, state and the city: Urban transformation in post-reform China. Habitat International 31(3): 303–316.

44.

Shen

Wong

Feng

(2002) State-sponsored and spontaneous urbanization in the Pearl River Delta of south China, 1980–1998. Urban Geography 23(7): 674–694.

45.

Smith

(2003) Remaking scale: Competition and cooperation in prenational and postnational Europe. In: Brenner

Jessop

Jones

, et al. (eds) State/Space: A Reader. Oxford: Blackwell Publishing, pp. 227–238.

46.

Solinger

(1991) The place of the central city in China’s economic reform: From hierarchy to network. City & Society 5(1): 23–39.

47.

Wong

Bird

(2008) China’s fiscal system: A work in progress. In: Brandt

Rawski

(eds) China’s Great Economic Transformation. New York: Cambridge University Press, pp. 429–466.

48.

(2002) China’s changing urban governance in the transition towards a more market-oriented economy. Urban Studies 39(7): 1071–1093.

49.

(2016) State dominance in urban redevelopment: Beyond gentrification in urban China. Urban Affairs Review 52(5): 631–658.

50.

Yeh

AGO

(2007) Urban Development in Post-Reform China: State, Market, Space. New York: Routledge.

51.

Yang

(1999) Urban-biased policies and rising income inequality in China. American Economic Review 89(2): 306–310.

52.

Yang

DYR

Chang

(2017) Financialising space through transferable development rights: Urban renewal, Taipei style. Urban Studies 55(9): 1943–1966.

53.

Yeh

AGO

Yang

Wang

(2015) Economic transition and urban transformation of China: The interplay of the state and the market. Urban Studies 52(15): 2822–2848.

54.

Yin

Liu

(2017) Administrative urbanization and city-making in post-reform China: A case study of Ordos City, Inner Mongolia. Chinese Geographical Science 27(6): 891–903.

55.

Zhang

(2006) China’s changing economic governance: Administrative annexation and the reorganization of local governments in the Yangtze River Delta. Regional Studies 40(1): 3–21.

56.

Zhang

(2008) Conceptualizing China’s urbanization under reforms. Habitat International 32(4): 452–470.

57.

Zhang

Zhao

(1998) Re-examining China’s ‘urban’ concept and the level of urbanization. The China Quarterly 154: 330–381.

58.

Zhou

(2012) Fiscal System and Behavior of Local Government. Shanghai: Sanlian Press (in Chinese).