To Build Outward or Upward: The Spatial Pattern of Urban Land Development in China

Transcription

1 To Build Outward or Upward: The Spatial Pattern of Urban Land Development in China Zhi Wang, Qinghua Zhang, Li-An Zhou September 5, 2016 Abstract This paper attempts to understand the patterns of urban land development in China from the political economy perspective. Our study is motivated by an interesting pattern observed in the past decade that in the initially more densely populated cities the urban land development was expanded more outwardly with a relatively low use intensity. We link this pattern with the career concerns of Chinese city leaders who pursue faster GDP growth and generate more fiscal revenues by expanding the city outwardly. We first develop a theoretical model to demonstrate how a city leader with career concerns chooses the spatial pattern of urban land development and derive several testable predications. First, city leaders with high career incentives are more likely to expand the city outward. The underlying mechanism that drives this is that by doing so the city leader can generate more land sale revenues for financing public infrastructure, which in turn enhances the economic performance and her career prospect. Secondly, the average use intensity of newly developed land is lower for city leaders with higher career incentive due to the existence of a mechanicaltrade-off between a city s upward and outward expansion.then we test these theoretical predictions using a large dataset on residential land transactions from 2000 to 2012 which are matched with the personal characteristics of city leaders during the same time period. We find robust evidence consistent with the predictions of our model. Our study highlights the important role of city leaders in shaping the unique patterns of urban land development in China. School of Economics, Fudan University, No. 600 Guoquan Road, Shanghai, , China. address: wangzhi@fudan.edu.cn, Phone: Guanghua School of Management, Peking University, Beijing, China, , address: Zhangq@gsm.pku.edu.cn, Phone: Guanghua School of Management and IEPR, Peking University, Beijing, China, address: zhoula@gsm.pku.edu.cn. 1

2 1. Introduction Land development is essential for urbanization because it will profoundly shape the urban internal structure and influence the spatial distribution of economic activities within a city. Such impact may even last for decades to come in the future. This is especially relevant for the fast-growing cities in the developing countries. Land development is subject to government regulations in most cities all over the world (Duranon and Puga, 2015; Gyourko and Molloy, 2015). It is thus very important to understand the key factors influencing the patterns of urban land development and regulations. China has experienced massive urbanization in the past two decades. According to the statistics from the National Bureau of Statistics of China, during , China s urbanization rate increased from 26% to 49.9%. In this process 372 million of people, a vast majority of whom were rural migrants, added into the urban population. Urbanbuilt-up land area also expanded substantially, increasing at an annual rate of 5.73%, or 134,550 hectares per year during the period of Such an astonishing scale of urbanization, which is unprecedented in the human history, has been accompanied by tremendous growth of real GDP in cities. In the past decade, China has revealed a very interesting pattern in urban land development. Figure 1 depicts the outward expansion of urban land development for China s 50 largest cities versus the citypopulation density in It is clear that initially more densely populated Chinese cities expand more outwardlyduring Figure 2 presents the use intensity of newly developed residential land reflected by the upper limits of regulatoryfloor-to-area ratio(far) versus population densityfor the same group of cities. It shows that initially more densely populated cities have lower FARlevels on average. This pattern is puzzling because for thosedensely populated cities in China such as Beijing, Shanghai, and Guangzhou, their developable land is usuallyscarcer relative to the size of the population they have to accommodate. On the contrary, manydensely populated cities in developedregions such as New York, London, Singapore, Tokyo, and Hong Kong tend to build more intensively. This paper studies the determinants of urban land developmentpatternsin China. City leaders play a central role in China s urban land development. On the one hand,they act as city developers and set up the key parameters shaping decisions about how much new land is to be developed and where to develop, and how to 2

3 regulate the intensity of land use such as the building density (FAR) for each land parcel. 1 This is very different from cities in the developed world where homeowners have considerable power in influencing local land development. On the other hand, Chinese city leaders who hold a critical control over local economic activities, are placed in growth tournaments (Li and Zhou, 2005; Xu, 2011) in which their promotion is closely linked with economic outcomes of the cities (such as growth of GDP and fiscal revenues). In this paper, we demonstrate that the career concern of city leaders is a key to driving the peculiar pattern of urban land development in China. There are very few studies in the literature examining the determinants of land development regulations in the developing world (Gyourko and Molloy, 2015). Most relevant studies are set in the context of developed world. Our paper fills in the gap. Also, the current literatureemphasizes the idea that incumbent home-owners have incentive to influence the local planners to restrict land development in their neighborhood in order to keep their property value from falling down (e.g., see Fischel, 2001; Glaeser, Gyourko and Saks,2005; Hilber and Robert-Nicoud, 2013; Ortalo-Magne and Prat, 2015). The influence from developers either does not show up in their frameworkor is not the focus of the game in the above literature. By contrast, our paper focuses on the decisions of city leaders who behave like grand developers as is the reality of China. Our paper thus offers an important complementary channel to the existing literature through which land development regulations are determined.we then utilize a large dataset that covers land transactions across 202 prefecture-level cities nationwide in China over the period of to conduct empirical analysis. Note that our data has rich cross-section and cross-time variations to help identify the channel that our theory proposes. We first develop a theoretical model to demonstrate how a city leader chooses the spatial pattern of urban land development. Specifically, she needs to decide both the urban boundary (i.e., how far the outward expansion of urban land development goes) and the land use intensity within the city reflected by certain regulatory FAR schedule. Her own career concern drives the decisions. The city leader's payoffs depend on her promotion likelihood, while this hinges on theproduct of her career concern intensity which is determined by the leader's personal conditions prior to her stepping in office, 1 Cai, Wang, Zhang (2016) and Brueckner, Fu, Gu, and Zhang (2016) show that the regulatory FAR are largely binding on residential development decisions in urban China. 3

4 and her ex-post performance during the term. If the city leader has higher career concern intensity, she will have stronger incentive to promote the city s economic performance since this can enhance her promotion likelihood and payoffs more effectively. City governments in China sell the long-term leaseholds of land use rights to private developers to conduct land development. The land sales revenues thus obtained are typically used to finance the urban infrastructure construction (Han and Kung, 2015) which in turn helps boost the local economic growth. Developing more land at the city edge (i.e., expanding the city more outwardly) can bring in more land sales revenues to the city treasury. However, the central government imposes control on the total amount of rural land to be converted to urban land. As a result, the city leader has to exert greater efforts to bargain with the upper level governments in order to obtain more quotas for rural-to-urban land conversion if she wants to expand the city more outwardly (Xie, 2015). Such efforts may be rather costly to the city leader. Only those city leaders with high career concern intensity are willing to exert efforts to expand the city boundary outwardly, because for them the benefits from increasing land sales revenues and boosting local GDP outweigh the effort costs. This is the first prediction of our model. The city leader also sets the density constraints of land use (i.e., regulatory FAR schedule) in the city. In doing so, she takes into account of the value of land considering various benefits and costs including the possible externality of high density development. Our theory shows that, if we hold housing demand factors such as income and population constant, then in equilibrium, the FAR schedule is lower as the city's boundary expands outwardly.this reflects a trade-off between a city's outward and upward expansion. Because city leaders with higher career concern intensity tends to expand city more outwardly, the second prediction of our model is city leaders with higher career concern intensity set lower regulatory FARs. Our empirical analysis utilizes the database on residential land transactions provided by the China Index Academy. The sample for our analysis contains detailed information (such as prices, sale date, and location) on over 30,000 completed residential land transactions (including residential mixed with commercial land) in 202 Chinese cities during We focus on residential land development because this type of land brings in around three quarters of total land sales revenues to local governments while land sale revenues is at the heart of our political economy 4

5 story about the urban land development. We also briefly discuss the development of industrial land and purely commercial land as corroborating evidence. The outward dimension of urban land development is usually measured in the currentliterature by the change intotal urban land area(e.g., BruecknerandFansler, 1983; McGrath, 2005; Deng et al., 2008; Lichtenberg and Ding, 2009). A clear limitation of the above measure is that the growth of urban land area may be caused by eitherdevelopments filling in open space among existing developments within the city boundary or those developments at the city edge that push the urban boundary further away from the city core. Only the latter is in essence the outward expansion of urban boundary. Taking advantage of our micro-level land data, we obtain the geographic coordinates for each land parcel fromwww.map.baidu.com. Using the coordinates, we calculate for each land parcel its distance to the city center. Wethen measure the extent of outward development during the term of each city leader with the top percentilesin the distribution of the land sdistance to the city center of all the land parcels sold during the city leader s term. We collected information on 974 city secretaries who were in office in the 202 Chinese cities contained in our land sample during The information includes the date of birth, education attainment, time when the term started, hierarchical level when the term started, time when the term ended, and status after the term ended (e.g., hierarchicallevel, retiring or not, position if not retiring, etc.). Based on each leader's personal conditions at the start of her or his office term, we are able to estimate the leader s prior chance of getting promoted at the end of the term regardless of her or his actual performance during the term. This measures the leader's career concern intensity. We then run regressions to check the relationship between a city leader's career concern intensity and the spatial pattern of urban land development. We find that a one-standard-deviation increase in the leader's career concern intensity will cause 11 km extra outward urban expansion, which is a 41% increase relative to the sample average.the key mechanism that drives this positive relationship between career incentives and outward expansion is a land-fiscal story. Specifically, through selling more land at the city edge, the city government can gather more land sales revenues, which can help finance the construction of urban infrastructure and stimulate economic growth. We provide some supporting evidence to the above mechanism. We find that an extra 10 km of average outward expansion of urban land will generate 5

6 extra land sales revenues of 10.6 billion, and increase total local GDP by 42.6 billion, which are 89.1% and18.6% compared to their sample averages respectively.we also show that outward expansion causes a decrease of FAR given other things equal, which is a mechanical trade-off between upward and outward dimension of urban development. We find that that a one-standard-deviation increase in the leader's career concern intensity will lower the FAR level by 0.042, which explains 12.4% of the variation of FAR across the terms of city leaders. Finally, for a more comprehensive view of urban land development, we discuss the pattern of both industrial land development and pure commercial land development based on a large transaction level data of these two types of land. We find that industrial land exhibits a similar outward expansion pattern to that of residential land; that is, higher-incentive city leaders develop more industrial land at the city edge. On the contrary, pure commercial land development has no such relationship with the local leaders career concern intensity. The rest of the paper is organized as follows. Section 2 discusses the related literature. Section 3 is a brief description of institutional background on land development as well as on local leaders' career incentives in China. In Section 4, we present a theory that demonstrates the decision making of city leaders regarding urban land development. Section 5 discusses data and main variables. We present our main empirical results regarding the outward expansion of residential land in Section 6. In Section 7, we show the empirical results regarding the FAR schedule of residential land. Section 8is a descriptive analysis of development patterns of two land use types (i.e., industrial and purely commercial) other than residential land. Finally, Section 9 concludes. 2. Related Literature Our paper is closely related to the literature on the determinants of urban land development regulations(see Gyourko and Molloy (2015) for a comprehensive review). A popular political economy story emphasizes the role of homeowners. Fischel (2001) argues that existing homeowners have a strong incentive to restrict any undesired land development in order to maintain their property value because their home is typically their primary asset. Glaeser, Gyourko and Saks (2005) claim that the homeowners' aversion to new land development contributes to the rising land costs in the U.S. since 1970s. Drawing on the same insight, Otralo-Magne and Prat(2014) 6

7 develop an OLG model that incorporates households location and housing investment decisions, as well as their collective decisions regarding housing supply regulations. Hilber and Robert-Nicoud (2013) provide some supporting evidence to the above story by showing that regions with a higher share of developed land relative to total available land tend to have more restrictive land development policies. However, the above studies are generally silent about or assign a passive role to developers in influencing the local land development policies. Molotch (1976) proposes the idea of "growth engine"; that is, people in charge of local governments tend to be local business leaders or similar types who naturally care more about economic activities. Therefore they are more likely to support land policies in favor of local economic growth. Logan and Molotch (1987) argue that urban growth machines are a typical feature of many U.S. cities. These growth machines are a combination of real estate, banking, and commercial interests who support the increased expansion of the city and are able to heavily influence urban policies. Fischel (2008) points out that developers have more influence in large governments. However, there have been few models in the literature providing a theoretical framework that focuses on the role of developers in the formation of land development policies. 2 Some empirical studies show preliminary evidence of the influence of developers. 3 Our paper contributes to the literature by explicitly studying how city leaders as grand developers play an important role in shaping the pattern of urban land development and setting FAR levels. Our model nicely aligns with the institutional background of China. In China, urban land development is under tight regulation mainly by the local government at the prefecture city level.the city government assumes responsibilities of both stimulating local economic growth and providing public goods in the city. Because the promotion of city leaders depends on the evaluation of their superiors which mainly is based on their economic performance, naturally, intense career concern will drive the city leaders to adopt land development policies that are more tilting towards local economic growth. On the other hand, the homeowners have relatively weak powers in affecting land use regulations in urban 2 Hilber and Robert-Nicoud (2013) and Glaeser, Gyourko and Saks (2005) incorporate the role of developers in their framework in a fairly simple way. 3 Sole-Olle and Viladecans-Marsal (2012) and Schone, Koch and Baumont (2013) finds some indirect empirical evidence pointing to the developers' influence on local land development regulations in Spain and in France, respectively. 7

8 China. Such background provides a nice context for us to develop a theory about, as well as to conduct an empirical test on, the influence of developers on urban land development, which may enrich the current literature. Career concern is a key factor that drives the decisions of city leaders in making land development policies. In this sense, our paper echoes with a vein of political economy literature on local leaders' behaviors and their promotion incentives in China. Li and Zhou (2005) show that this evaluation system had traditionally placed heavy emphasis on economic growth statistics. Gong, Knight, Xiao and Zhang (2016) find that promotion incentive drives fiscal expenditure cycles for Chinese cities. By investigating the data for 86 Chinese cities during , Zheng, Kahn, Sun, Luo (2014) show that there exists an association between environmental performance and an urban leader s probability of being promoted. Kahn, Li, and Zhao (2015) find that younger local officials who have greater career concerns are more enthusiastic about attaining the green target laid down by the central government. Our paper also relates to the local public finance literature. One frequently discussed motive for land use regulations in the literature is fiscal externality on the provision of local public goods (e.g.epple, Romer and Filimon, 1988; Calabrese, Epple and Romano, 2007). More recently, Henderson and Venable (2009) demonstrate thatrevenues from land developing play an important role in the dynamic formation of cities. Han and Kung (2015) shows that fiscal pressures tend to push the urban land area to grow more in China because the local governments need to grab more fiscal revenues from land sales. Our paper further enriches this literature by demonstrating that city governments raise land sales revenues through land development to finance public infrastructure and enhance local economic growth. Because different city leaders have different career concern intensity, they put different weight on economic performance when designing various land development policies, which leads to variations in the pattern of urban development across cities and over time. Finally, this paper is broadly related to a small yet growing literature on land development in urban China. Deng, Huang, Rozelle and Uchida (2008) and Lichtenberg and Ding (2009) study various economic factors that influence the urban land growth in China. Zheng, Fu, Liu (2006) and Zheng and Kahn (2008) examine the housing and land markets within a Chinese city and estimate the land price gradient and FAR gradient. Brueckner, Fu, Gu and Zhang (2016) study the stringency of FAR constraints in China, while Cai, Wang and Zhang (2016) investigate the 8

9 implementation of FAR regulations in Chinese cities. None of the above paper addresses how city leaders make decisions regarding urban land development policies. 3. Institutional Background 3.1 Regulations on urban land development in China The 1980s and early 1990s had witnessed an unprecedented loss of cultivable land due to large-scale industrialization and urbanization. The sizable decline in cultivable land intensified the central concerns about China s food security, 4 and in the meantime the conflicts between local governments and discontent farmers due to the insufficient compensations for the land conversion have beenintensified sharply. In order to keep urban land expansion under control, the central government amended the Land Administrative Law in 1998 andenacted a new set of cultivable land protection provisions aiming at prohibiting any additional loss of cultivated land. In the same year, a new ministry, the Ministry of Land and Resources of the People s Republic of China, was established to strengthen the central government s planning control over land development. Since 1998, anurban land quota systemhas beenimplemented via a hierarchical, top-down planning process. The central government makes the nation s long-term plan of land development. The first plan covered the period between 1997 and The plan dictates the maximum amount of newly developed urban land for each province in the long run as well as the minimum amount of cultivable rural land in reserve. These are two important constraints. Accordingly, provincial governments make their own long-term plans of land development that allocate land use quotas to cities under their administrative control. In addition, the governments at various levels also make short-term (5-year or annual) land development plans which provide more detailed guidelines. The city governments make decisions regarding the total amount of land and the location of land to be developed in a city. When a development involves conversion of rural land to urban land, it must get approval from the upper level government. If the conversion is within the boundary of quota constrains imposed by the upper government, then it is relatively easier to be approved. Otherwise, the city 4 At the beginning of the economic reforms, in 1978, the cultivated rural land per capita in China is only 1.55 mu, far below that of the U.S. 9

10 government has to exert a lot of efforts to lobby the upper level government and/or the central government for approval. 5 In the conversion of rural to urban land, the city government has to compensate the farmers who are currently using the land. By China s Land Administration Law, the compensation fee (all together) should not exceed 30 timesthe average annual value of product generated from the land within the three years before the conversion. 6 This is typically far below the market value of land which can be 500 times higher than the compensation fees (Liu, Chen and Zhao, 2012; Zhang, 2007). Naturally, this causes the discontent of farmers. However, in most cases, they are weak political power and have to obey the rules set by the local governments.it is thus clear developing rural land at the city edge is lucrative to the city governments since they can auction off the land at much higher price after conversion. If a city leader has strong career incentive, she will be much attracted to expand the city more outwardly since this can bring more fiscal revenues which can finance public infrastructure and improve her economic performance. In developing more rural land for urban use at the city edge, the city governments has various ways to bypass the quota constrains set by the upper level governments (Xie, 2015). They typically figure out legitimate reasons to persuade the upper governments to grant extra land use quota after their short term quota has been used up. Such reasons may include building a college town at the city edge, building a cultural and historic town or developing a high-tech industrial park. As a result, the quota of the long-term plan has been almost used up years before the end of the plan. For example, in 2004, Shandong and Zhejiang provinces had used up 80% and 99% of the total land quota of the long term plan of , respectively. The zeal of local governments for urban land placed tremendous pressure on the central government when making the next new long-term land development plan. Meanwhile, the city governments actively lobby and bargain with the upper level governments for favorable treatment in the next long-term plan, which may effectively legalize their current urban expansion beyond the quota constraints set by the old plan. In addition, sometimes, the city government even tolerates some small scale land development at 5 China s Land Administration Law (1998) requires that any conversion of farm land to nonagricultural use must be approved from higher level authorities and must be offset by conversion or reclamation of other land to agricultural use so that amount of land in agriculture, adjusted for quality, remains constant (Ding, 2003; Lichtenberg and Ding, 2008).. 6 The Law also says that in some specific cases, the compensation fee can be increased more than 30 times the upper bound. However, the local governments are generally unwilling to do so. 10

11 the city edge without any approval from the upper governments. In general, going beyond the quota constraints is possible, but at considerable costs, such as lengthy bargaining, proposing creative rationales for extra quota, and networking with upper level government officials. After rural land is converted to urban land, it is at the disposal of the city government. By law, all urban land is owned by the state.since 1988, the use rights of vacant urban land have been allocated through leaseholds by each city s land bureau. In the 1990s, most use rights allocations were done by negotiation between developers and government officials. To control widespread corruption in such negotiated land deals, in 2002 the Ministry of National Land and Resources banned negotiated sales after August 31, Since then, all urban leasehold sales for private development have been conducted through public auctions. In each city, land auctions are held by the local land bureau, with details of all transactions posted to the public on the Internet. All land sales revenues are turned into the city treasury. Now let us turn to regulations on the intensity of land use. The city's land reserve and allocation committee conducts land use planning and sets the general guidelines for regulations on land use with the aims of promoting rational land use, guarding public interests, and protecting historic heritage and natural resources. This committee consists of the city s political leaders and key figures from relevant local government agencies, such as the land bureau and the urban planning bureau. Following those guidelines, the city urban planning bureau then determines use type and detailed development restrictions (e.g., regulatory FAR limit, building height, green area rate, etc.) before each land parcel is released to the land bureau for auction.when setting the detailed land use restrictions, the planning bureau aims tomaximize the value of landconsidering various benefits and costs including the possible externality such as the costs rising from high density development. The FAR regulation is one of the most important land use regulations in urban China. By law, any land parcel to be auctioned off must have a designated regulatory FAR level. Also, after the land is developed, the city s planning bureau must complete an official inspection of the residential project before it is put up for sale, in order to ensure compliance with the FAR regulation. In most cases, the FAR regulation takes the form of an upper bound constraint on the ratio of a building s total floor area to the lot size on which the building is to be constructed. As Cai, Wang and Zhang (2016) shows, although the FAR constraints can be adjusted under certain circumstances, 11

12 they are largely binding in urban China. Therefore, FAR regulations play an important role in shaping the urban spatial pattern of Chinese cities. We use the regulatory FAR upper limits to measure the land use intensity, which, intuitively, can be thought of as the upward dimension of urban land development. 3.2 Local leaders career concern and land-fiscal policy The Communist Party of China (hereinafter referred to as CPC) is the single ruling party in China. The Central Committee of the CPC has the ultimate power over personnel within the system. There are two important officials at each layer of administration: the Party Secretary, a figure that represents CPC, and the head of local government. The political leader at each level is the Party Secretary. The local party secretary is appointed by their superiors in CPC. Prefecture cities are under the supervision of provinces. In China's centralized personnel system, it is the upper level committee of the Communist Party who determines the promotion of local leaders, while economic performance is the most important criterion for the superior to evaluate local leaders (Li and Zhou, 2005). Another critical factor in a local leader s career is her age and hierarchical level. The retirement age is set according to the local leader s hierarchical level. And it is enforced rather strictly in China. Considering her current age and administrative level, she can anticipate how far away she is from the glass ceiling of retirement. The further away she is from the glass ceiling, the higher career concern intensity she has. If on the contrary, she is about to hit the ceiling, she will have no much incentive to work hard to improve the local economic performance. Since the late 1970s, China has gone through several rounds of fiscal reforms in an effort to decentralize its fiscal system and fiscal management (Zhang and Zou, 1998). Since the 1994 tax reform, China established a streamlined tax system geared to thesocialist market economy. Under the current tax system, there are 18 types of taxes in China, divided into central taxes, local taxes, and taxes shared by the central and local governments. Taxes which are related to national priorities, such as security, are classified as central taxes, and these include the consumption tax. Taxes related to local development are classified as local taxes, and these include the business tax, individual income tax, and the property tax. Other tax revenues are divided between the central and local governments, and these include the value-added tax and the enterprise income tax. Local taxes and shared taxes constitute a large portion of in-budget revenue of local 12

13 government. However, since only the central government has the authority to make tax laws and tax policy, local governments have little control over these in-budget revenues. Under the highly centralized political system in China, city leaders respond to promotion incentives by stimulating economic growth. Stimulus often requires revenues;however, after the tax reform in 1994, the in-budget fiscal revenues of local governments shrank tremendously. City governments have thus turned to off-budget revenues, such as land leasing fees and debts, to finance urban public goods provision such as education, pensions, unemployment insurance, housing and infrastructure investment (World Bank, 2012), which in turn can help boost the local GDP growth. Available evidence suggests that the ratio of land sales revenues to local fiscal revenues of prefecture city governments increased from 10% in 1999 to nearly 60% in 2005 (Han and Kung, 2015). Note, about three quarters of land sales revenues come from the sale of residential land. The city governments tend to deliberately lower the sale prices of industrial land in order to attract the entry of new firms which again may enhance the GDP growth. 4. The Model In this section, we present a theoretical framework to demonstrate how city leaders make decisions regarding urban land development, from which we derive testable predictions. Assume a mono-centric city with linear form. All workers commute to CBD to work. A. Individual An individual worker s utility comes from the housing services, in addition to the consumption of a composite good of all other commodities. The person s objective function is as follows: max uxr ((), hr ()) xr () hr () xr ( ), hr ( ) st.. w x() r h() r p() r 2tr, (1) whererindicates the location where the person lives, x(r) is consumption of the composite good of the person, w is wage income, tis commuting cost per unit distance, and p(r) is unit floor area house price at location r. Solving the individual utility maximization and considering the location 1 13

14 equilibrium condition, we have (1 ) ( w 2 tr) pr () V, (2) wherev is individual worker s indirect utility and (1 )( w 2 tr) hr (). pr () (3) B. Firm Assume the city's production is conducted by a representative firm. The firm's production function is Y AGL, (4) wherea is technology, G is public infrastructure provided by the government, and L is labor. 7 In eq. (4),G can improve the total factor productivity. We assume the firm sells product in the national market and there is free trade among cities. The product price is normalized to be 1. Prefect competition among firms gives zero profit, which implies that: w AG. (5) Eq. (5) pins down the wage rate. Suppose each worker provides 1 unit of labor. When the labor market clears, L=N, where N is city s population size. C. Build Upward: floor-to-area ratio Informed by the institutional background in urban China as discussed in Section 3, we assume the city government delegates some agent to set the floor-to-area ratio Fr () for land development at each location r. In making the decision, the agent aims to maximize the value per unit of land; that is, Fr ( ) arg max prf ( ) qff ( ) (6) F The construction cost per floor area is qf ( ). The construction cost qf ( ) increases monotonically with F. For example, foundation costs may increase with the height of buildings. These costs may also include expenditures incurred in correcting the negative externalities such as noise and traffic jam rising from high density development. Note, thedecision of FAR by the agent is similar to what a private 7 For simplicity, we do not include capital in the production function because the focus of this paper is not on capital. One can easily add in capital assuming infinite elastic supply of capital to the city from the national capital market. This would complicate the model solution without gaining further insights. 14

15 developer would do (DiPasquale and Wheaton, 1996), except for that the city government considers the possible externality costs. Let the construction cost per unit floor area be qf ( ) F, 0. Then from (6), the optimal FAR level is p( r) F( r). (7) 2 D. Outward expansion: benchmark case where the city government is not a developer Let us first look at thebenchmark case where it is not the city government, but the market force that drives the urban outward expansion. In this case, the city government does not assume the role of a developer. The private developer obtains land from its current user at the market price, denoted as p l (.).The developer chooses how much land to develop in order to maximize profits. The urban boundary S is thus determined by the following condition: p( SFS ) ( ) qfs ( ( )) FS ( ) p( S) p (8) The left hand side of the above equation is the marginal revenues of urban land development at the city edge. The right hand side is the market price of land at the city edge, which reflects the opportunity cost of converting agricultural land to urban use. In equilibrium, this equality must hold. Suppose the land market structure is such that in equilibrium, the unit land price is close to the marginal revenues from developing the land; that is, 15 l p () r p() r F() r q( F()) r F() r (9) l The city government collects land sales tax at a ratet. The tax revenues are all used to finance urban public infrastructure G. Thenthe government s budget constraint is S S S 0 l 0 0 G t p () r dr t p() r F() r dr q( F()) r F() r dr (10) Also note in equilibrium, free migration across cities implies that V v (11) where V is the individual s indirect utility. And the housing market clearance implies S Fr () dr hr () N. (12) 0 Combining the above four equilibrium conditions (9)-(12) with equations (2), (3), (5)

16 and (7), we can solve for the equilibrium for this benchmark case. Let us denote the equilibrium urban boundary in this benchmark case as S 0. E. Outward expansion: real life case where the city government assumes the role of developer Now let us turn to the case where the city government behaves as a developer and sets the urban boundary. In making the decision regarding urban boundary, a city leader concerns her own pay-offs which derives fromthe success of her political career. If promoted, the city leader s payoff is K 0, otherwise, her payoff is zero. Because more fiscal revenues can help her political career, the city leader has incentive to grab more revenues from land development. All the revenues from land development are collected by the city government. The net revenues from land development are equal to the total house value subtracting the land compensation cost and construction cost. 8 Compensation costs are unavoidable in land development. Such costs typically include costs of demolishing existing structures and compensating current residents. In our model, we use a continuous function, c(r), to measure the compensation costs. Developing land near the CBD usually involve much higher compensation costs than developing land far away from the CBD (Zhu, 2004). Thus, we assume that cr () decreases with r. The revenues from land development are all used to finance urban public infrastructure, G, which will improve the city s total factor productivity and boost GDP growth. Thus, the city government s budget constraint is S S S G prfrdr () () qfr ( ()) Frdr () crdr () 0 0 (13) 0 To convert farm land to urban land, the city government only has to pay the farmers a compensation cost that is far belowthe market value of land due to various institutional reasons (see Section 3.1). 9 Thus at the previous benchmark city edge S 0, the LHS of equation (8) which is the marginal revenues from land development is greater than the compensation costs. This will induce the city government who longs for more revenues to expand the city boundary further than S 0. 8 In urban China, the city government is the monopoly supplier of city land. It holds auction to sell long-term leaseholds of land to private developers. Through competitive bidding, the land sales revenues can grab the developers profits from land development. 9 The government revenues here is typically much higher than the tax revenues in the bench mark case where the city government does not assume the role of a developer. 16

17 Specifically, the decision process is as follows. Let us use the city s total output Y to capture economic performance in the model. We use a parameter to capture a city leader s career concern intensity which depends on her personal characteristics such as age and hierarchical level when stepping in office. The higher is, the further away is the local leader from the career glass ceiling, so the leader concerns her career more intensely. As such, we define the city leader s promotion likelihood function as f ( Y ), with f '(.) 0. So the leader s promotion likelihood is an increasing function of the product of her career concern intensity and her ex post economic performance. 10 When the career concern intensity is higher, improving performance can increase the leader s chance of promotion more effectively. The city leader faces a trade-off. On the one hand, expanding the city outward can bring more fiscal revenues so as to finance public infrastructure and accommodate more people in the city, which in turn boosts GDP growth. On the other hand, pushingthe urban boundary outward usually requires the city leader to seek more urban land quotas from the upper level government which is costly. Because the central government is aware of the fact that the costs of obtaining land at the city edge by the city governmentsare much lower than the market value of land. As such, it makes long-term plan of urban land development which sets the quota of rural-to-urban land conversion for each province according to projected population growth and economic growth. The provincial government who is the direct upper level government of the city government then allocates the quota across cities and over time. Such quota constraints are reflected by constraints on city s boundary S in our model. Suppose the provincial government sets the city boundary constraint at the benchmark urban boundary S If a city leader wants to expand the spatial boundary of the city beyond S 0, she needs to make extra efforts in lobbying the upper level governments, in order to obtain extra quota for converting rural land to urban land. In this sense, such boundary constraints are not strictly binding and subject to costly bargaining. As described in section 3, Chinese city leaders use several ways in practice to seek more quota and 10 In the western political regimes, politicians efforts to win votes tend to decrease with their wining chances (e.g., Besley, Persson and Sturm, 2010; Sole-Olle and Viladecans-Marsal, 2012). In contrast, in China s political system, the likelihood of promotion of a politician hinges on both the politician s personal conditions prior to her stepping in office and her ex-post performance during the term. The two factors are complementary to each other. 11 In our simulation, we suppose that the city boundary constraint is set by the upper level government at the benchmark level of the market equilibrium. Our model s main predictions are robust to other choices of S 0 17

18 push the urban boundary outwardly. The lobbying efforts for more urban land quota are reflected in the effort cost function, denoted as E( SS ; 0). We assume that these lobbying costs increase with S when the city expansion goes beyond the boundary constraint S 0. Specifically, we define the effort costs of the city leader as 0, if S S0 ESS ( ; 0 ), (14) e ( S S0), if S S0 where 0, 0, and e 0.. The city leader's objective function is described as follows: max f( AGN) K E( S; S ) S st.. 0 () i G p() r F() r dr q( F()) r F() r dr c() r dr. (15) ( ii) V v ( iii) S Fr () dr N 0 hr () S S S In (11), (i) is the city government s budget constraint,(ii) is the intercity spatial equilibrium condition under free migration, and (iii)means that the total housing supply should accommodate the total housing demand. E. Solution and simulation Plugging equations (2), (3),(5), (7), and (13) into the objective function (15), we can derive the first order conditions for S. With this first order conditions plus three constraints conditions, we can solve for 4 unknown variables, i.e., S, G, N, and V. They fully characterizethe equilibrium. Since analytical solution is impossible to obtain, we use reasonable parameters to conduct simulations. From simulations, we can do comparative statics and demonstrate how urban land development patterns are determined. Figure 3 shows that stronger career concerns drive a city leader to expand the city boundary more outwardly. Figure 4 helps us further understand the mechanism underlying such urban development patterns. It shows a positive relationship between the city's spatial boundary and several economic outcomes in our model.the more outward a city expands, the more fiscal revenues, more populationand higher total output. These will enhance the city leader s chance of getting promoted. In addition, the urban outward expansion has an implication on building upward 18

19 intensity; i.e. the floor to area ratio. As a city s urban boundary expands outwardly, if we keep the wage and population of the city unchanged, individual people can have more housing consumption. This implies the individual s utility level increases. According to eq. (2), the housing price at each location declines. As a result, the FAR level at each location decreases too, according to eq. (7). This reflects the mechanical trade-off between urban outward and upward expansion. In sum, we have the following testable predictions. First, city leaders with high career incentives are more likely to expand the city outward. The underlying mechanism that drives this is that by doing so the city leader can generate more land sale revenues for financing public infrastructure, which in turn enhances the economic performance and her career prospect. Secondly, holding the wage and population of the city unchanged, the use intensity of newly developed land is lower for city leaders with higher career incentive due to the existence of a mechanical trade-off between a city s upward and outward expansion. 5. Data and Variable Construction 5.1 Constructing measures for the spatial pattern of urban land development To characterize the spatial pattern of urban land development, we exploit the database on residential land transactions provided by the China Index Academy, China s largest independent think tank focusing on real estate market. The sample for analysis contains information on over 30,000 completed residential land transactions in 202 Chinese cities during For each land parcel, we know its use type, land area, regulatory FAR, reserve price, selling price, transaction date, auction type, etc. As for the outward urban development, we are essentially interested in measuring the extent by which a city s edge of urban land developments is pushed farther away from the city center. Taking advantage of the micro-level land data, we are able to do so with detailed geographic locations of the city s residential land developments. Specifically, we obtain the geographic coordinates for each land parcel 12 Among a total of 202 cities in the sample, 198 are prefecture-level cities, and 4 are provincial-level cities (i.e., Beijing, Shanghai, Tianjin, and Chongqing). 13 The original sample has 36,058 residential land parcels with completed sales. 5,229 land parcels are dropped because their regulatory information (e.g., FAR) is missing. An additional 11 land parcels are dropped because their geographic location information is not available. Finally, 801 land parcels are dropped because the information on city leaders is missing for these observations. The final study sample contains 30,017 residential land parcels. 19

20 from Using the coordinates, we calculate for each land parcel the distance to the city center. 14 Panel B of Table 1 shows the summary statistics of land s distance to the city center by the time period of land transaction and by region (i.e., coastal, central, northeast, southwest, northwest). More than half of the residential land transactions in China happened in the coastal regions. For the period , the average distance to the city center of residential land development is about 18 km in the coastal region and drops to around 10 km for the rest of the country. That is, on average, urban residential developments are located further away from the city centers in the coastal cities than other cities, probably due to their relatively large historical urban areas. Over time, as more land parcels in the suburbs were developed, the edge of urban land developments has been pushed outward. In the full sample, the 90 percentile in the distribution of land s distance to the city center is 35.7 km in the period and increases to 63.2 km in the period To investigate the relationship between outward development and local leader s career incentives, we measure the extent of outward development during the term of each city leader with the top percentiles in the distribution of the distance to the city center among the land parcels sold during the city leader s term. To study the effect of outward development on local economic growth, we use the mean distance to the city center among land parcels sold in each city in each year. 15 We have a total of 871 city/year-level observations and 386 city/secretary-level observations for regression analysis. The outward expansion of urban land developments is a typical form of urban spatial expansion, which the previous studies usually measure with the change of urban land area using various data sets (e.g., BruecknerandFansler, 1983; McGrath, 2005; Deng et al., 2008; Lichtenberg and Ding, 2009). A clear limitation of the aggregate measures is that the growth of urban land may be caused by either new developments filling in open space among existing developments or those pushing the fringe of the existing urban development further away from the city center, while only the latter is consistent with the outward developmentexpansion of urban land 14 We use the coordinates of the 1992 light center (i.e., the brightest cell at night in each city s central area) from Baum-Snow et al. (2016) to identify the actual city center. They suggest that despite the enormous increase of light over the past two decades, the light centers have remained unchanged. 15 The effect of noises increases as the study time period gets shortened, and the top percentiles are more likely to suffer from the measurement errors due to the noises than the mean. Therefore, we choose to focus on the mean distance when the study time period is year. Note that, on average, each term lasts for 4 years. 20