Air Quality in Urban China

Air Quality in Urban China Canfei He, Teng Zhang, and Wang Rui1

Abstract: This study develops a framework for measuring the capability, pressure, and resistance to enforce environmental regulations and then applies this three-dimensional enforcement framework to air quality enforcement in urban China. Based on daily Air Pollution Index data published by China’s Ministry of Environmental Protection during 2001–2011, the paper reports an overall improvement in urban air quality, but significant inter-city variations. S tatistical results indicate that low capability and strong resistance to the enforcement of stringent environmental regulations have contributed to environmental degradation in urban China, and social pressure has not significantly helped clean the urban sky. Economic development and structural transformation have tended to reduce the environmental bargaining power of regulated actors in local economies. The findings imply that supportive institutional arrangements together with necessary financial and human resources would help improve air quality. J ournal of Economic Literature, Classification Numbers: Q200, Q500, Q530. 5 figures, 4 tables, 55 references. Key words: air pollution index, environmental regulation, enforcement, panel data regression, environmental Kuznets curve, urban air quality.

INTRODUCTION

C

hina’s exceptional economic performance has come at the cost of environmental degradation (Economy, 2004; He et al., 2012). Air pollution in urban China is particularly severe, and has become a global concern, with Economy (2004) reporting that the country now has 16 out of the world’s 20 most polluted cities. Anti-pollution protests successfully halted work on the PX projects2 in Xiamen and Dalian, and in 2012, environmental protests in Qidong of Jiangsu Province and Shifang of Sichuan evolved into serious episodes of social unrest. The pro-growth, resource-intensive, and export-oriented development strategy that China has pursued, coupled with the institutional relationships designed to support this strategy, have certainly posed a serious problem for environmental protection efforts (Jahiel, 1997, 1998; van Rooij and Lo, 2010; He et al., 2012). Empirical studies have investigated air pollution across Chinese provinces based on the framework of the environmental Kuznets curve (EKC) (Shen and Hashimoto, 2004; Shen, 2006; Liu et al., 2007; Poon et al., 2006; He, 2006, 2009, 2010; Song et al., 2008). This line of research suggests that pollution is inevitable in the early stages of economic development, but that there will be alleviation as the economy reaches a certain turning point. The Working Paper series of the World Bank and related publications have examined the influence 1 Respectively, College of Urban and Environmental Sciences, Peking University, Beijing, 100871 China and Peking University-Lincoln Institute Center for Urban Development and Land Policy ([email protected]. cn); School of Urban Planning and Design, Peking University, Beijing, 100871 China; and Department of Urban Planning, School of Public Affairs, University of California at Los Angeles, Los Angeles, CA 90095. The authors acknowledge funding from the Natural Science Foundation of China (41271130, 40830747, 41071075). 2 PX is an abbreviation for paraxylene, a chemical used to make polyester products (for details, see Hao, 2011).

750 Eurasian Geography and Economics, 2012, 53, No. 6, pp. 750–771. http://dx.doi.org/10.2747/1539-7216.53.6.750 Copyright © 2012 by Bellwether Publishing, Ltd. All rights reserved.

he et al. 751

of pollution charges, environmental subsidies, ownership structure, and the bargaining power of factories and community pressure on the emission behaviors of industrial polluters in China (Wang and Chen, 1999; Wang, 2000, 2002; Wang and Wheeler, 1996, 2000, 2005; Wang et al., 2003; Dasgupta et al., 2001; Wang and Jin, 2002, 2007). He et al.(2012) argued that marketization and decentralization have been harmful to the urban environment, while globalization has been beneficial to urban air quality. These studies have certainly advanced our understanding of environmental degradation in China. Realizing the importance of environmental protection, since the early 1980s China has gradually instituted a system of environmental regulations meant to control and prevent pollution. The Chinese government today pays more attention to the environment than was the norm in virtually all Western countries at comparable stages of their economic development (Smil, 1998). While the legislation is thought to have improved, the effect of law is limited, largely because it is insufficiently and ineffectively enforced (Ma and Ortolano, 2000; van Rooij and Lo, 2010). The gap between environmental regulation and regulatory enforcement is especially pronounced in China. The root of the enforcement problems lies in the institutional framework, in which the local governments fund and directly manage the local environmental authorities (Sinkule and Ortolano, 1995; Tang, et al., 1997; Ma and Ortolano, 2000; van Rooij, 2002, 2006; van Rooij and Lo, 2010). An emerging literature on policy implementation in China has improved our understanding of its complexity. Policy implementation in China relies on an administrative system characterized by an authoritarian, top-down structure and a decentralized administrative system. Environmental regulations are mostly formulated at the central level and enforced by local officials (Jahiel, 1998; Zhan et al., 2009; van Rooij, 2010). Enforcement of environmental regulations in China is conditioned by a variety of cultural, institutional, organizational, and contextual factors. Substantial regional variation exists in the enforcement of environmental regulations in China (Wang and Wheeler, 1996; Schwartz, 2003; van Rooij, 2010). Schwartz (2003) identified the factors of state capacity, government commitment, and public participation to explain differences in enforcement of environment regulations, while van Rooij (2010) stressed the importance of central-level policy, community pressure, local government commitment, organizational capacity, regulated actors, and economic conditions. We anticipate that geographical differences in these factors will influence air pollution levels in Chinese cities. This study seeks to empirically test the impacts of enforcement of environmental regulations on air quality. Theoretically, this study will develop a framework that incorporates three dimensions of regulatory enforcement—i.e., the capability to enforce, the pressure to enforce, and the resistance to enforcement. It then employs the best proxies to quantify these three dimensions of enforcement and test their significance using a panel data regression model. Based on daily Air Pollution Index (API) data published during 2001–2011, this study reports an overall trend of improving air quality in urban China, but with significant intercity variations. Statistical results suggest that the capability to enforce stringent environmental regulations results in better air quality. Strong resistance to enforcement of stringent environmental regulations, however, would impede environmental protection efforts. The remainder of this paper is organized as follows. The second part consists of a literature review and proposes a theoretical framework to relate enforcement of environmental regulations to air pollution. It is followed by a third section, which describes the data sources employed and the spatial and temporal patterns of air quality in urban China. The fourth section presents the model specification and the statistical results. The paper concludes with a summary of empirical findings and discussion of policy implications.

752

EURASIAN GEOGRAPHY AND ECONOMICS

ENFORCEMENT OF ENVIRONMENTAL REGULATIONS AND AIR QUALITY IN URBAN CHINA: CONCEPTUAL FRAMEWORK AND HYPOTHESES The EKC literature proposes an inverse U-shaped relationship between economic development and pollution (Grossman and Krueger, 1991). A number of empirical studies confirm the presence of EKC for some pollutants in China (Shen and Hashimoto, 2004; Shen, 2006; Liu et al., 2007; Song et al., 2008; Poon et al., 2006; He, 2006, 2009, 2010; He et al., 2012). Economic growth has both harmful and beneficial effects on environmental quality. The harmful effects are associated with scale effects and the beneficial effects accrue from structural and technological effects (Grossman and Krueger, 1991). The institutional relationship concerning environmental regulations can shape the relationship between economic development and pollution (Panayotou, 1997; He et al., 2012). Since the early 1980s, China has gradually implemented a system of environmental laws and regulations to prevent pollution (Jahiel, 1997). Regulatory enforcement in China, however, depends on “three legs”—the central government, local governments, and a broad array of external watchdogs/monitors (Economy, 2005). Yet in reality China’s central government has limited capacity to inspect implementation activities at the local level; local governments can easily impede law enforcement; and environmental NGOs are still relatively weak in terms of policy advocacy capacity and holding local governments accountable (Tang and Zhan, 2008). In the current political framework, primary responsibility for environmental quality falls to local environmental protection bureaus (EPBs). Local EPBs must ensure that factories install pollution prevention technology, operate waste treatment facilities, reduce harmful emissions, or pay fees if these emissions exceed standards. Local EPBs must ensure that city planning agencies reject proposals from polluting firms seeking to invest in the area. Accordingly, local EPBs must have sufficient authority to carry out their responsibilities (Jahiel, 1998). However, EPBs have historically been weak institutions with limited authority and resources (Jahiel, 1998; Lo and Leung, 2000; Ma and Ortolano, 2000). China’s decentralized administrative system means that the local governments provide EPBs with their annual budgetary funds, approve institutional advancements in rank, and determine increases in personnel and even the allocation of such resources as cars and office buildings (Jahiel, 1998). The difficulties of implementation and enforcement are rooted in local governments protecting their local economies, environmental protection agencies lacking authority and resources, and ignorance of environmental policy staff with regard to the law and legal procedures (Tang et al., 2003; van Rooij, 2003). EPBs at various levels have found it especially difficult to obtain the active support and cooperation of other bureaucratic authorities in charge of economic development in order to take a tough stand on environmental problems (Chan et al., 1993). With their strong pro-growth orientation, government agencies such as planning, economic, and construction commissions and industrial and commercial authorities are reluctant to endorse and enforce stringent environmental measures to protect local economic interests (Sinkule and Ortolano, 1995; Jahiel, 1997, 1998; Ma and Ortolano, 2000; Tang et al., 2003). Furthermore, EPBs lack authority, administrative rank, and financial and human resources (Chan et al., 1993; Jahiel, 1997, 1998; Tang et al., 1997; Ma and Ortolano, 2000). Local EPBs suffer from weak internal management processes (van Rooij, 2003, 2006) as well as the riskaverse nature of their enforcement agents, who are afraid to upset powerful regulated actors or other government officials (van Rooij, 2006). These problems are exacerbated when EPBs do not receive support from local communities and environmental groups (Dasgupta et al., 2001;

he et al. 753

Fig. 1. Enforcement of environmental regulations and air quality in urban China.

Lo and Fryxell, 2005) or when they face powerful enterprises that try their best to undermine enforcement actions (van Rooij, 2006). In certain cases, China’s legislative process may proceed quickly on certain issues and typically involve less consultation with sectoral interests (Lo, 1995). National legislation may also lack local legitimacy: local actors may not share the environmental concerns of national regulation (Jahiel, 1998; van Rooij, 2002). Very often, local governments develop their own environmental policy priorities to guide enforcement in order to minimize the adverse impact on local economic growth. In this way, local EPBs are under substantial pressure to ensure that enforcement will not contradict the local development strategy (Jahiel, 1997; Ma and Ortolano, 2000). In China, characterized by an authoritarian, top-down structure and a decentralized administrative system, systemic institutional arrangements may undermine the enforcement of environmental regulations. Substantial regional variation in the enforcement of environmental regulations in China has been observed, which can be accounted for by a combination of institutional factors (Schwartz, 2003; van Rooij and Lo, 2010). A number of studies have examined factors influencing EPB enforcement actions, including the severity of local pollution, the structure of ownership and financial solvency of firms, and public complaints lodged against polluting firms (Wang et al., 2003). This study develops a framework to that incorporates the capability to enforce environmental regulations as well as the pressure for and resistance to such enforcement (Fig. 1). The following sections will analyze how the triple-dimension of enforcement influences air quality in urban China. Capability to Enforce Environmental regulations are set by the central government and implemented by local governments in China. Despite the mass of laws and policy tools in place in China, it seems that compliance with regulations is poor (MacBean, 2007; Zhan et al., 2009). Beyond the national institutional root cause undermining the enforcement of environmental regulations,

754


the capability of local governments to enforce may differ and would influence air quality in urban China. The regulatory enforcement demands inputs of financial and human resources. To improve the capability of environmental protection, local governments have invested more in it, recently providing stronger support for local EPBs (van Rooij and Lo, 2010). For instance, in the State Environmental Protection Model Cities (including the coastal cities of Dalian, Zhuhai, and Xiamen), local governments boast strong environmental reputations matched by environmental spending and support. Even traditionally pro-growth cities such as Guangzhou, Wuhan, and Chengdu have become more environmentally friendly (Lo and Fryxell, 2005; Lo and Tang, 2006). However, the enforcement gap in environmental regulations in China nonetheless is often due to lack of capability. Local EPBs have chronic shortages of staff and materials such as cars, which are needed for inspections. Understaffed and increasingly overburdened EPBs are often unable to carry out regular, proactive inspections at all of the polluting enterprises (van Rooij, 2003). Implementation of environmental impact assessment and the assessment of discharge fees are impeded by inadequate technology and finances, limited human resources, faulty data, and inferior agency reports. The low discharge fees themselves reduce their effectiveness as a deterrent (Swanson et al., 2001). van Rooij (2003) reported that the interior province of Sichuan does not perform as well as the more affluent provinces in East China, mainly due to a lack of local government funding and local environmental awareness. As shown in a number of case studies, China’s local environmental management system is seriously underfunded and is often faced with strong pro-growth opposition within the local party-state establishment (Jahiel, 1997; Tang et al., 1997; Lo et al., 2000). The central government has mandated that local governments pay for crucial social services such as education, healthcare, and social welfare, while providing no guarantee that funds would be made available for localities that lack the resources to do so, leaving many local governments in very poor financial shape (Wong, 1995, 2009). Local governments typically accord environmental protection a low investment priority when facing fiscal hardship. However, some studies suggest that strengthening the capacity of local EPBs, i.e., increasing their funding and improving their human resources, can help overcome the obstacles to regulatory enforcement (Dasgupta and Wheeler, 1996; Ma and Ortolano, 2000). Sufficient capability to enforce would promote the effective and efficient enforcement of environmental regulations at the local level and help reduce air pollution in urban China. Consequently, the first research hypothesis is: H1: Strong capability to enforce stringent environmental regulations results in better air quality. Pressure to Enforce Severe environmental degradation caused by China’s remarkable economic growth has motivated the central government to pay more attention to environmental protection. Meanwhile, with increasing income, local residents care more about environment protection. Local governments therefore come under pressure to enforce environmental regulations from both upper-level governments and local residents. First, pressure from upper-level governments may come from the environmental performance of local governments through enforcement campaigns (van Rooij, 2006) and the environmental quality administrative leadership accountability system introduced in 1997 (Lo and Tang, 2006). In enforcement campaigns, central governmental agencies cooperate

he et al. 755

to organize a national effort to detect and punish certain violations of pollution laws (van Rooij, 2002; Economy, 2004). During such campaigns, local EPBs are found to increase the number of inspections and punish violations more aggressively (van Rooij, 2002). Variation of central-level pressure on regions to enforce environmental regulations more strictly can be detected, inasmuch as central pressure for tighter regulatory enforcement has been far greater on coastal regions than inland areas. For instance, the Pearl River Delta Region (Lo and Fryxell, 2005; Lo and Tang, 2006; van Rooij, 2010) and the Huai River Region (Economy, 2004), both notorious for lax enforcement, were prioritized by the central authorities for extra scrutiny during campaign-driven actions. Central pressure should force local governments to implement more stringent environmental regulations. Second, mounting public pressure through petitions, collective actions, and social conflicts over environmental pollution has prompted local government officials to take the environment much more seriously. As the Chinese have become more affluent and better educated, they have started to demand a better living environment and pay more attention to environmental protection. Social pressure has been found to influence the enforcement of environmental regulations. Environmental complaints have significantly increased the numbers of inspections and sanctions (Dasgupta et al., 2001; van Rooij and Lo, 2010), and have contributed to pollution reduction especially for air (Dasgupta and Wheeler, 1996). The incidence of citizen complaints is strongly affected by educational level and per capita income (Dasgupta and Wheeler, 1996; Wang and Wheeler, 2003). Plants located in areas of higher GDP per capita are experiencing stricter environmental regulation. Environmental agencies do take the public interest in a plant’s pollution into account when they determine plant regulation. Environmental regulation is the outcome of balancing all (environmental) costs and (economic) benefits into account. Community complaints can help EPBs pressed for resources inside the bureaucracy (van Rooij and Lo, 2010). For instance, the Guangzhou EPB was able to act on increasing public complaints against vehicular pollution to convince the municipal government to adopt a total ban on nonleaded gasoline in the 1990s (Lo and Leung, 2000). Thus, although social pressures can help overcome EPB capacity deficiencies, there is variation in how social pressure and support affects enforcement of environmental regulations.3 There may be a shift in enforcement strategy from more reactive to more proactive. However, social pressure no longer directly drives enforcement, because local governments no longer largely depend on the public for detection of violations (van Rooij, 2006). Thus, theoretically the net impact of pressure to enforce is unclear. However, when citizens can use new legal and political possibilities awarded by central-level commitment to protect the environment, the combination of public and central-government pressure can force local governments to implement stringent environmental regulations that should help improve air quality in urban China. The second research hypothesis is thus: H2: Pressure to enforce environment regulations helps to improve air quality. Resistance to Enforcement In many policy arenas, support from and collaboration with key stakeholders is crucial for effective policy implementation. Environmental policy implementation especially requires cooperation from stakeholders, such as policymakers, industrial bureaus, regulated businesses, 3 Lo et al. (2006) reported that social support actually had a negative impact on enforcement effectiveness in Dalian and Guangzhou.

756


environmental NGOs, and community groups (Jahiel, 1998; Imperial, 2005; Thomas, 2003). Recently, external stakeholders such as private firms, civil society organizations, and others have played a more important role in China’s environmental governance (Economy, 2004, Tang and Zhan, 2008). The resistance to environmental enforcement in China mainly comes from bureaucrats charged with promoting economic development (Lieberthal, 1997; Ma and Ortolan, 2000) as well as the firms subject to regulation (Wang and Jin, 2002; van Rooij and Lo, 2010). With a strong pro-growth strategy, local governments have usually sided with the economic bureaus and have subordinated environmental protection to economic interests whenever the two have been seen to be in serious conflict (Lieberthal, 1997; van Rooij, 2002). Environmental agencies at various levels have found it difficult to obtain the active support and cooperation of other bureaucratic authorities in charge of development. Powerful government agencies in charge of economic and urban development are known to be quite reluctant to endorse and enforce stringent environmental measures for fear that they might slow economic growth (Ma and Ortolan, 2000). There are numerous examples in which economic, planning, trade, and industry bureaus have collectively conspired to undermine the strict enforcement of environmental regulations, particularly in the enforcement of environmental impact assessment requirements (Lo et al., 2000; Ma and Ortolano, 2000). Many powerful bureaucratic authorities either directly own or sponsor polluting enterprises or development projects with adverse environmental consequences. Consequently, the enforcement of environmental regulations often has a great and unwelcome impact on the domains and jurisdictions of various government bureaus. Thus, lack of institutional support has thus made EPBs quite hesitant to enforce stringent environmental regulations. The bargaining power of Chinese firms in enforcing environmental regulations depends on the type of ownership and the level of public pressure (Wang and Wheeler, 2003). For instance, Wang and Wheeler (2003) found that effective assessment rates are sensitive to regional ambient air quality, local incidence of pollution-related complaints, factory profitability, ownership, production, sales, and economic sector. In general, large firms contributing substantial tax revenue and value-added or dominant employers or those owned by governments at various levels are more difficult to regulate. If one or a few large firms dominate a city’s economy, their leaders may be able to exert greater leverage on city officials. In the past, local EPBs have had to enter difficult negotiations with state-owned enterprises to get them to comply with the law. SOEs enjoy stronger bargaining power with local governments regarding environmental regulations than do privately owned enterprises (Wang and Jin, 2002). SOEs are often closely connected with local governments and some managers of SOEs have higher political status than the local environmental authorities. As a result, SOEs are able to elicit a lower pollution punishment and, as a result, are less likely to decrease their air pollution. Using plant-level data, Wang and Wheeler (2000) found that SOEs are more likely to pollute than private enterprises, which is further confirmed in a study by Wang and Jin (2002) and He et al. (2012). The changing structure of local economies (in the direction of private ownership) thus may reduce the bargaining power of dominant firms. Enforcement becomes easier when the economy becomes more heterogeneous, with fewer dominant employers and a clearer separation between the state and enterprises. But enforcement may be more difficult as the number of enterprises increases, with more small enterprises seeking every possible means to cut costs due to increasing competition (van Rooij and Lo, 2010). Overall, we believe that resistance to enforcement of environmental regulations at the local level should have negative impacts on air quality. Therefore, the final research hypothesis is:

he et al. 757

Fig. 2. Chinese cities publishing daily API in different years. Number of stations added to the network is indcated in parentheses after each date, 2000–2011.

H3: Resistance to enforcement of environmental regulations results in poor air quality. The following session of the paper will first describe the temporal and spatial pattern of air quality using the daily published air pollution index (API) and then test the significance of the proposed research hypothesis using a panel data regression model. TEMPORAL AND SPATIAL PATTERN OF AIR QUALITY IN URBAN CHINA Air Pollution Index The Ministry of Environmental Protection is responsible for measuring the level of air quality in China. Starting on June 5, 2000, China began monitoring the daily pollution level in 42 of its major cities; 5 and 37 cities were added in June 5, 2001 and June 4, 2004, respectively. By 2011, about 120 cities were publishing a daily air pollution index (API). The spatial distribution of the monitored cities is shown in Figure 2. They include major cities located throughout the nation and are representative of Chinese cities as a whole. For instance, in 2000, the 42 cities included all provincial capitals and other key cities such as Dalian, Qingdao, Shenzhen, Nantong, Suzhou, Wenzhou, Xiamen, Yantai, Zhanjiang, and Zhuhai. The API level is based

758


Table 1. Limiting Values for Individual Pollutants in API API

Pollutants (μg m–3) SO2 (daily)

NO2 (daily)

PM10 (daily)

CO (hourly)

O3 (hourly)

50

0.05

0.08

0.05

5

0.12

100

0.15

0.12

0.15

10

0.20

200

0.80

0.28

0.35

60

0.40

300

1.60

0.56

0.42

90

0.80

400

2.10

0.75

0.50

120

1.00

500

2.62

0.94

0.60

150

1.20

on the level of five atmospheric pollutants, namely sulfur dioxide (SO2), nitrogen dioxide (NO2), suspended particulates (PM10), carbon monoxide (CO), and ozone (O3) measured at monitoring stations scattered throughout each city. An individual score is assigned to the level of each pollutant and the final API is the highest of those 5 scores. The pollutants can be measured quite differently. SO2, NO2, and PM10 concentrations are measured as average per day. CO and O3 are more harmful and are measured as average per hour. The evaluation of the pollution level for pollutant p (PIp) is carried out by a linear interpolation of the reference scale values reported in Table 1 PI hi – PI lo PI p = ------------------------------  C p – BP lo  + PI lo BP hi – BP lo

where PIp is the value of the PI for pollutant p, Cp is the daily reference concentration of pollutant p, BPhi is the lowest breakpoint of pollutant p that is greater than or equal to Cp, BPlo is the highest breakpoint of pollutant p that is less than or equal to Cp, PIhi is the PI value corresponding to BPhi of pollutant p, and PIlo is the PI value corresponding to BPlo of pollutant p. The API is then derived as follows, API = max{PISO2, PINO2, PIPM10, PICO, PIO3} Based on API, the MEP classify seven categories of air quality: Excellent (0–50), Good (51– 100), Slightly Polluted (101–150), Lightly Polluted (151–200), Moderately Polluted (200– 250), Heavily Polluted (250–300) and Severely Polluted (>300). API has been published daily for the monitored cities with a maximum limiting value of 500. This study will explore the temporal and spatial variations of air quality on an annual basis. Because the API limiting value is 500, we would expect that the median will be a better indicator to illustrate the annual average air quality for individual cities. We also look at the 10 percentile and 90 percentile API values and number of days with API greater than 150. TEMPORAL AND SPATIAL CHANGES OF AIR QUALITY IN URBAN CHINA To show the overall trend of air quality in urban China, we computed the mean and median APIs and the number of days with API greater than 150 for 42 cities that have published daily API during the period 2001–2011. We also simply use the full sample of cities

he et al. 759

Fig. 3. Overall trend of air quality in urban China, 2001–2011.

to calculate the same set of indicators. Both are shown in Figure 3. The statistics show that air quality in urban areas has consistently improved over the past decade. For those 42 cities reporting API values throughout the entire period of study, the annual median API decreased from 78.81 in 2001 to 63.36 in 2011, and the number of days with API greater than 150 fell substantially, from 24.95 days to 5.23 days. Correspondingly, the number of days with excellent air quality (relative to the total number of days) increased from 16.22 percent in 2001 to 25.99 in 2011, and the number of days with slightly polluted to severely polluted levels (relative to the total number) fell from 6.84 to 1.52 percent. The full sample exhibits a generally similar trend. However, there is substantial city variation within the overall trend of air quality change. Figure 4 depicts the changing pattern of air quality during 2001–2011 in selected cities. Air quality in cities such as Dalian, Suzhou, Xiamen, Guangzhou, and Kunming are relatively stable, particularly since 2006, with only a little fluctuation. For instance, the Median API for Dalian is 61 in 2001 and 59 in 2011. Kunming recorded a median API of 57 in both years. Some cities with poor air quality, such as Beijing, Shijiazhuang, Taiyuan, Chongqing, and Lanzhou, have exhibited decreasing API. Lanzhou and Shijiazhuang for instance had a median API of 125 and 126, respectively, in 2001. But in 2011, the two median APIs fell to 83 and 71, respectively. Beijing had been preparing for the 2008 Olympic Games since 2001 and the efforts have paid off. During 2001–2011, the median API in Beijing decreased from 100 to 75. Not unexpectedly, some cities evidenced increasing environmental degradation. The coastal cities of Ningbo and Wenzhou exhibited a rising API, indicating deteriorating air quality. Guilin in Guangxi, a famous tourism city, has also observed increasing API. The air quality in the capital city of Xinjiang, Urumqi, has also deteriorated since 2003. The different patterns of change in air quality over time in urban China may be associated with a set of economic, geographic, and institutional factors. For instance, the coastal cities have been pursuing a heavily industrialization strategy, attracting chemical and heavy industries and generating more air pollution. Corresponding to the different temporal trends of air quality in urban China, there is considerable intercity variation in air quality. Figure 5 portrays the spatial distribution of median API and the 10th and 90th percentile API at the city level in 2006 and 2011. There

760


Fig. 4. Sample cities with relatively stable air quality (top), deteriorating air quality (middle), and improving air quality (bottom).

are salient geographical patterns. First, coastal cities have better air quality than those in the interior. Many coastal cities have median APIs smaller than 60. Second, cities north of the Yangtze River have poorer air quality than those in the south. Third, large cities, including

he et al. 761

Fig. 5. Spatial distribution of air quality in 2006 (86 cities) and 2011 (120 cities).

the provincial capitals, observe relatively poor air quality. Finally, however, intercity variations remain in the inland and northern regions: some inland and northern cities report smaller median APIs than others. The 90th percentile API shows a more polluted pattern, in which some southern cities are also heavily polluted. The 10th percentile API is similar to the median API. Although regional and physical factors contribute to the spatial pattern of air quality in

762


urban China, there are many other undermining factors, which would shed light on the level of air quality. The identification of institutional factors influencing air quality is the focus of this study. EMPIRICAL ANALYSIS AND RESULTS Variables and Models To understand the intercity variation of air quality in China and to test the proposed research hypothesis, this study systematically investigates air quality by applying a panel data regression model for the period 2006–2010. The particular focus is to explore the impact of enforcement of environmental regulations, and hence the explanatory variables include proxies for capability to enforce, pressure to enforce, and resistance to enforcement, controlling for some physical factors. The main dependent variable is the median API (API_50), with API at 10% (API_10) and 90% (API_90) as additional dependent variables. The first set of variables proxy for capability to enforce, which is associated with urban size, economic development, and human resources. We include urban population (POP) and GDP per capita (PGDP) to quantify the overall capability of enforcement. Size matters in dealing with environmental protection. Positively speaking, large cities are able to command sufficient resources to protect their environment and clean their air. However, air pollution is also highly concentrated in large cities. The relationship between economic development and capability to enforce may be nonlinear, meaning that only at a certain stage of development are local governments willing to enforce stringent environmental policies. We thus introduce the squared per capita GDP (SPGDP) in the model. The introduction of per capita and squared per capita is often used to test the EKC effect (He, 2006, 2009; He et al., 2012). The negative impact of squared per capita GDP on air pollution is attributed to structural and technological effects, which can be proxies for capability of enforcement. We expect a negative coefficient on SPGDP.4 The fiscal situation is particularly related to the capability of enforcement, because local governments are responsible for the budget of local EPBs and provide EPBs necessary resources to implement environmental regulations (Jathiel, 1998). We employ the ratio of local expenditure to local revenue (FISC) to measure the fiscal situation. A high value for this ratio indicates a better fiscal situation and stronger capability of enforcement. FISC is lagged by one year in the model specifications. A negative coefficient is expected for FISC. Finally, local environmental protection agencies conduct many different activities and the implementation gap is often due to lack of personnel. We introduce a variable to measure the number of people in environmental law implementation in local EPBs per 10,000 persons (EPB). This number is only available at the provincial level. We apply the number to all cities within the same province. More people in law implementation suggest strong capability to enforce. We anticipate EPB has a negative coefficient on API. Pressure to enforce emanates from two sources—the central government and local residents. The central government has paid increasing attention to environment protection and has exerted pressure on local governments. However, it is difficult to directly quantify the pressure. The study period is 2006–2010, during which the 11th Five-Year Plan was implemented. In this plan, the central government set an energy saving and emissions reduction target for each province. We use this target (EGOAL) to measure central government pressure. Cities In other words, the larger the value of SPGDP, the lower the value of API_50 (the better the air quality).

4

he et al. 763

in the same province are assumed to experience the same central pressure.5 Social pressure arises from local residents and is associated with economic development and level of education. People with higher income and educational levels tend to care more about the air quality. We introduced the number of environmental complaints per 10,000 persons (CMPL) and the number of college students per 10,000 persons (COLL) as proxies for social pressure. CMPL is lagged by one year in the models. For CMPL, only provincial-level data are available, so we applied the provincial average to cities located in the same province. Both variables are expected to have negative coefficients. In addition, some cities are striving to develop their tourism industries, which requires good air quality and thus may impose some pressure to enforce stringent environmental regulations. Therefore, we introduced a dummy variable for key national tourist cities (TOUR) and expect a negative coefficient. Local governments encounter resistance to the enforcement of stringent environmental regulations. Such resistance, from the unsupportive institutions, can be related to a poor fiscal situation, which would induce strong resistance from the governmental authorities in charge of economic development. We introduced two variables to quantify resistance from regulated firms. State-owned enterprises (SOEs) have strong connections with governmental agencies and are constantly found to have stronger bargaining power than other types of enterprises (Wang and Wheeler, 2000; Wang and Jin, 2002). Local EPBs may face strong resistance to enforcement of environmental regulations in SOE-dominated local economies. In some less-developed and small and medium-sized cities, a few firms may dominate the economies in terms of employment, tax revenues, and value-added. Local governments often heavily protect those key contributors. Such dominance consequently would grant these enterprises strong bargaining power with local governments regarding environmental regulations. We introduce the percentage of SOEs’ output in total industrial output (SOE) and the percentage of the three largest industrial firms’ output (TOP3) to quantify the resistance to enforcement. Strong resistance would result in poorer air quality in urban China. Finally, air quality is influenced by a variety of physical factors. In the model specification, we control for the altitude, annual precipitation, average annual wind speed, and the annual average temperature. The definition of variables is summarized in Table 2. Empirical Results Correlation analysis was used to investigate the relationship between the dependent and independent variables. Table 3 reports the correlation coefficients. The three dependent variables are significantly and positively correlated, but with moderate coefficients, justifying the application of three dependent variables. The correlation coefficient between API_10 and API_50 is 0.87, while that between API_10 and API_90 is 0.61. Proxies for resistance to enforcement have the expected associations with the dependent variables. For instance SOE and TOP3 are positively associated with API. Some variables for capability and pressure to enforce do not share the expected correlations with API, indicating that their roles are apparently conditioned by other factors. With respect to coefficients among explanatory variables, SOE and TOP3 are moderately correlated, with a coefficient of 0.7031, which is expected inasmuch as SOEs are typically large firms in China. The two provincial-level variables EPB and EGOAL are also moderately related, with a coefficient of 0.68. Other associations are fairly weak. Given the moderate and weak associations among explanatory variables, we do not see a serious collinearity issue in the model estimations. However, the central government may in fact set higher goals for cities that are heavily polluted.

5

764


Table 2. Definitions of Variables and Expected Signs Variable Capability to enforce

Pressure to enforce

Resistance to enforce Controlling variables

Definitions

Expected sign

POP

Urban population (10,000 persons)

?

PGDP

Per capita GDP (10,000 yuan)

?

FISC

Ratio of local expenditure to local revenue

–

EPB

Number of people in environmental law enforcement in local EPBs per 10,000 persons

–

EGOAL

Energy saving and emissions reduction target in Chinese province in the 11th Five-Year Plan

?

CMPL

Number of environmental complaints per 10,000 persons

–

COLL

Number of college students per 10,000 persons

–

TOUR

Dummy variable for tourist cities

–

SOE

Percentage of SOEs’ output in total industrial output

0

TOP3

Percentage of three largest industrial firms’ output

0

ALTITUDE

City altitude (masl)

RAIN

Annual precipitation (in 0.1 mm)

WIND

Annual average wind speed (in 0.1 m/s)

TEMP

Annual average temperature (in 0.1 °C)

Estimations of the time fixed effect model, the cross-sectional model, and the random effect model for the three dependent variables are presented in Table 4. All model estimations are highly significant and the explanatory power is fairly high. The results indicate that physical conditions do influence air quality in urban China. Cities with more rain, higher mean wind velocity, and higher temperature are typically cleaner. In the fixed and random effect models, the dummies for years are also significantly negative, confirming the trend that urban China has been improving its air quality. Given the significance of physical conditions, one can still observe the importance of economic and institutional factors for improving air quality in urban China. Urban size does matter for air quality. The variable lnPOP has a positive and significant coefficient, indicating that large cities are typically associated with poor air quality. Large cities are certainly able to provide the necessary resources to clean their air, but the concentration of people, economic, and transportation activities may have offset the environmental protection effort. As a proxy for overall capability of enforcement, PGDP and its square are barely significant in the models of API_50 and API_90 regardless the estimation methods. Once controlling for physical and institutional factors, economic development does not appear to relate to air quality, especially in heavily polluted cities. The EKC effect is only observed in the cross-sectional model of API_10. The EKC literature stresses the importance of technological and structural transformations. However, we can interpret that technological progress and structural transformation are subsumed within the category of capability of enforcement. Some studies interpret that economic development is manifest in social pressure to enforce stringent environmental regulations (Wang and Wheeler, 2000). People with higher income will demand better living

0.3767* 0.3133* 0.2244* –0.0729 –0.0457 0.0477 1

EPB

EGOAL CMPL

0.0406 –0.1683* 1

COLL

TOUR

0.2121* 0.1108 –0.1603* –0.022

0.0366 0.7031* 1

TOP3

lnALT

RAIN

–0.1812* –0.0814 –0.3399* –0.0338 1

WIND

TEMP

a

* p F Prob chi2 N

–2

API_10

–1

API_10

Table 4. Regression Results for API_10, API_50, and API_90a

766 EURASIAN GEOGRAPHY AND ECONOMICS

he et al. 767

environments and bring pressure to bear on local governments to improve air quality. This income-related pressure likely occurs in environmentally friendly cities. Economic development may not have worked effectively to curb air pollution. However, we do find some evidence that the capability to enforce environmental regulations is associated with air quality. In the models of API_50 and API_90, FISC has an expected negative coefficient, which is significant in the cross-sectional estimations. Fiscal improvement would allow local governments to invest more in environmental protection and enforce stringent environmental regulations. Those efforts would result in better air quality. Personnel in environmental law and regulation implementation also affect air quality in urban China. EPB has a negative and significant coefficient in the fixed effect models, indicating that the more people engaged in environmental law enforcement the lower the API. Adequate staffing may well guarantee better enforcement of stringent environmental regulations, as shortages of personnel are often cited as the cause of the environmental implementation gap in China. Our results in this regard are consistent with the findings in van Rooij (2003), van Rooij and Lo (2010), Schwartz (2003), and Tang et al. (2003). The results thus provide evidence in support of our first research hypothesis (H1). Systematically, pressure to enforce environmental regulations has not impacted air quality yet. EGOAL, representing pressure from the central government, has a positive and significant coefficient in the random effects model of API_50 and API_90, suggesting that the central government has set higher energy-saving and emission-reduction goals for polluted regions. Most studies argue that the root of enforcement problems lies with the institutional arrangement in which local governments pay and directly manage China’s main local environmental enforcement authorities (Tang et al., 2003; Tilt, 2007; van Rooij and Lo, 2010). Yet in reality China’s central and provincial governments have limited capacity to inspect implementation activities and outcomes at the local level (Economy, 2005). The central government often launches political campaigns, which would tend to increase both the number of inspections carried out and the punishment of violators during the campaign period (van Rooij, 2006). However, this type of political campaign may not impose continuous pressure on local governments and thus not achieve the long-term goal of improving the environment. The insignificance of central pressure to promote enforcement has the institutional and technical causes. Social pressure has not reduced air pollution either. Controlling for other variables, CMPL does not have a significant coefficient, implying that more environment related complaints have not translated into pressure to enforce stringent environmental regulation and bring about better air quality. Surprisingly, COLL has a positive coefficient and is significant for API_50 and API_90. Air quality may be poorer in cities with more well educated people and better tertiary education. In fact, these cities are typically super-cities including Beijing, Shanghai, Wuhan, Guangzhou, Xi’an, Guangzhou, and Shenzhen. The concentration of people and increasing dependence on cars has worsened air quality in the larger cities recently. It may also be argued that well-educated people in China’s large cities are willing to sacrifice living environments for better personal opportunities currently and may not express their environmental concern and exert pressure on local governments. TOUR has a negative but insignificant coefficient in the models of API_10. Overall, the empirical findings do not support the second research hypothesis (H2). Social pressure has not significantly helped to improve air quality in urban China. Although pressure to enforce environmental regulations has not improved air quality, resistance to enforcement may have worked against environmental protection efforts in urban China. SOE has expected positive and significant coefficients in most model specifications.

768


The positive association between SOEs and industrial pollution has been reported in several studies including Wang and Wheeler (2000), Wang and Jin (2002), and He et al. (2012). The findings indicate that cities dominated by SOEs have poor air quality. SOEs have substantial bargaining power and strong connections with local and upper-level governmental agencies, and are able to resist the enforcement of stringent environmental regulations, resulting in poor air quality. With inclusion of SOEs, TOP3 turns negative but is only significant in the cross sectional model of API_50. Often the cities’ largest enterprises are the targets of local environmental complaints; they care about their reputation in the community and are eager to maintain good relations with local government. Thus, they would make some effort to reduce pollution emissions. Meanwhile, it also suggests that if the top three largest industrial firms are not SOEs, the air quality may be better. The third research hypothesis (H3) is thus at least partially supported. SUMMARY AND DISCUSSION China has consistently been a focus of worldwide attention, for both its rapidly growing economy and the serious environmental degradation that has occurred within its borders since the 1980s. Economic transition has not only liberalized the Chinese economy but also presented a serious challenge to environment protection efforts. Air pollution in urban China has become a global concern. Numerous studies have stressed the critical role played by structural and technological factors in air pollution. However, in a country characterized by an authoritarian, top-down structure and a decentralized administrative system, the root cause of environmental problems is not simply economic or technological, but political and institutional. Complementary to the existing literature, this study traces the gap in the implementation of environmental regulations to air pollution and provides an institutional understanding of air quality issues in urban China. Specifically, this study examines the gap of environmental implementation as a function of the capability to enforce environmental regulations, public and central-government pressure to do so, and the resistance of regulated enterprises to such enforcement. Based on the daily API data during the period 2001–2011, this study reports an overall trend of improving air quality in urban China, but with significant intercity variations. Comparatively, cities in interior and Northern China have poor air quality. Statistical results suggest that the capability to enforce stringent environmental regulations and policies is positively associated with air quality. A strong capability to enforce environmental policies should improve air quality. On the contrary, resistance to enforce stringent environmental regulations would work against environmental protection efforts and result in poor air quality. Although there is tremendous pressure from upper-level governments and the community at large for local governments to protect the environment, this pressure has not significantly affected air quality in urban China. The enforcement of environmental regulations in China is a complicated process, involving governments at the central and local levels and many external stakeholders. Given the current institutional framework, the regulatory bodies have strong incentive to avoid enforcement of environmental regulations whereas the enterprises that are supposed to be regulated have substantial bargaining power to elicit reduced punishments for pollution. Our findings imply that the strengthening of supportive institutional arrangements, together with the necessary financial and human resources, will help improve urban air quality in China. As the Chinese economy becomes more structurally diversified and liberalized, large individual enterprises will become less dominant, and their environmental bargaining power will shrink

he et al. 769

accordingly. Meanwhile, as the Chinese people become more affluent and better educated (and increasingly environmentally aware), social pressures will rise and force local governments to take greater action to protect the environment. As a consequence, institutional improvements accompanying social, structural, and technological developments should make urban China’s skies a cleaner in the coming years. REFERENCES Chan, H. S., K. C. Cheung, and J. M. K. Lo, “The Socio-political Limits of Environmental Control in the People’s Republic of China: A Case Study in Guangzhou in Public Administration,” in S. Nagel and M. Mills, eds., Public Policy and the People’s Republic of China. New York, NY: Macmillan, 1993, 63–82. Dasgupta, S., B. Laplante, N. Mamingi, and H. Wang, “Inspections, Pollution Prices, and Environmental Performance: Evidence from China,” Ecological Economics, 36, 487–498, 2001. Dasgupta, S. and D. Wheeler, Citizens Complaints as Environmental Indicators: Evidence from China. Washington, DC: World Bank, Policy Research Working Paper Series 1704, 1996. Economy, Elizabeth, The River Runs Black: The Environmental Challenge to China’s Future. Ithaca, NY: Cornell University Press, 2004. Economy, Elizabeth, “Environmental Enforcement in China,” in Kristen A. Day, ed., China’s Environment and the Challenge of Sustainable Development. New York, NY: M. E. Sharpe, 2005, 102–120. Grossman, G. and A. Krueger, Environmental Impacts of a North American Free Trade Agreement. Cambridge, MA: National Bureau of Economic Research, Working Paper No. 3914, 1991. He, Canfei, Fenghua Pan, and Yan Yan, “Is Economic Transition Harmful to China’s Urban Environment? Evidence from Industrial Air Pollution in Chinese Cities,” Urban Studies, 2012 [doi: 10.1177/0042098011415719]. He, J., “Pollution Haven Hypothesis and Environmental Impacts of Foreign Direct Investment: The Case of Industrial Emission of Sulfur Dioxide (SO2) in Chinese Provinces,” Ecological Economics, 60, 228–245, 2006. He, J., “China’s Industrial SO2 Emissions and its Economic Determinants: EKC’s Reduced vs. Structural Model and the Role of International Trade,” Environment and Development Economics, 14, 2:227–262, 2009. He, J., “What is the Role of Openness for China’s Aggregate Industrial SO2 emission? A Structural Analysis Based on the Divisia Decomposition Method,” Ecological Economics, 69, 868–886, 2010. Hao, Tang, “Public Storm in Dalian,” Chinadialogue, September 9, 2011 [http://www.chinadialogue .net/article/show/single/en/4511-Public-storm-in-Dalian], accessed October 19, 2012. Imperial, Mark, “Using Collaboration as a Governance Strategy: Lessons from Six Watershed Management Programs,” Administration & Society, 37, 3:281–320, 2005 Jahiel, A., “The Contradictory Impact of Reform on Environmental Protection in China,” China Quarterly, 149:81–103, 1997. Jahiel, A., “The Organization of Environmental Protection in China,” China Quarterly, 156:757–787, 1998. Lieberthal, Kenneth, “China’s Governing System and Its Impact on Environmental Policy Implementation,” China Environmental Series, 1:3–8 (Washington, DC: The Woodrow Wilson Center), 1997. Liu, L.-C., Y. Fan, G. Wu, and Y.-M. Wei, “Using LMDI Method to Analyze the Change of China’s Industrial CO2 Emissions From Final Fuel Use: An Empirical Analysis,” Energy Policy, 35, 11:5891–5900, 2007. Lo, C. W. H., “Environmental Protection in Hong Kong amidst Political Transition: Is Hong Kong Ready to Manage Its Environment by Law?,” Environmental Management, 19, 3:331–344, 1995.

770


Lo, C. W. and G. E. Fryxell, “Governmental and Societal Support for Environmental Enforcement in China: An Empirical Study in Guangzhou,” Journal of Development Studies, 41, 4:558–588, 2005. Lo, C. W. H. and S. W. Leung, “Environmental Agency and Public Opinion in Guangzhou: The Limits of a Popular Approach to Environmental Governance,” The China Quarterly, 163:677–704, 2000. Lo, C.W. and S. Tang, “Institutional Reform, Economic Changes, and Local Environmental Management in China: The Case of Guangdong Province,” Environmental Politics, 15, 2:190–211, 2006. Lo, C. W. H., P. K. T. Yip, and K. C. Cheung, “The Regulatory Style of Environmental Governance in China: The Case of EIA Regulation in Shanghai,” Public Administration and Development, 20, 3:305–318, 2000. Ma, X. and L. Ortolano, Environmental Regulation in China. Lanham, MD: Rowman & Littlefield Publishing Group, 2000. MacBean, A., “China’s Environment: Problems and Policies,” The World Economy, 30, 2:195–362, 2007. Panayotou, T., “Demystifying the Environmental Kuznets Curve: Turning a Black Box into a Policy Tool,” Environment and Development Economics, 2, 4:465–484, 1997. Poon, J., I. Casas, and C. He, “The Impact of Energy, Transport, and Trade on Air Pollution in China,” Eurasian Geography and Economics, 47, 5:568–584, 2006. Schwartz, J., “The Impact of State Capacity on Enforcement of Environmental Policies: The Case of China,” Journal of Environment and Development, 12, 1:50–81, 2003. Shen, J., “A Simultaneous Estimation of Environmental Kuznets Curve: Evidence from China,”, China Economic Review, 17, 4:383–394, 2006. Shen, J. Y. and Y. Hashimoto, Environmental Kuznets Curve on Country Level: Evidence from China. Osaka, Japan: Osaka University, Graduate School of Economics and Osaka School of International Public Policy (OSIPP), Discussion Papers in Economics and Business 04–09, 2004. Sinkule, B. and L. Ortolano, Implementing Environmental Policy in China. Westport, CT: Praeger, 1995. Smil, Vaclav, “China’s Energy and Resource Uses: Continuity and Change,” The China Quarterly, 156: 935–951, 1998. Song, T., T. Zheng, and L. Tong, “An Empirical Test of the Environmental Kuznets Curve in China: A Panel Cointegration Approach,” China Economic Review, 19, 3:381–392, 2008. Swanson, K. E., R. G. Kuhn, and W. Xu, “Environmental Policy Implementation in Rural China: A Case Study of Yuhang, Zhejiang,” Environmental Management, 27, 4:481–491, 2001. Tang, S., C. W. Lo, and G. E. Fryxell, “Enforcement Styles, Organizational Commitment, and Enforcement Effectiveness: An Empirical Study of Local Environmental Protection Officials in Urban China,” Environment and Planning A, 35, 1:75–94, 2003. Tang, S., C. W. H. Lo, J. M. K. Lo, and K. C. Cheung, “Institutional Constraints on Environmental Management in Urban China: Environmental Impact Assessment in Guangzhou and Shanghai,” The China Quarterly, 152:863–874, 1997. Tang, Shui-Yan and Xueyong Zhan, “Civic Environmental NGOs, Civil Society, and Democratization in China.” Journal of Development Studies, 44, 3:425–448, 2008. Thomas, Craig W., Bureaucratic Landscapes: Interagency Cooperation and the Preservation of Biodiversity. Cambridge, MA: MIT Press, 2003. Tilt, B., “The Political Ecology of Pollution Enforcement in China: A Case from Sichuan’s Rural Industrial Sector,” The China Quarterly, 192:915–932. 2007. van Rooij, B., “Implementing Chinese Environmental Law through Enforcement: The Shiwu Xiao and Shuangge Dabiao Campaigns,” in J. Chen, Y. Li, and J. M. Otto, eds., The Implementation of Law in the People’s Republic of China. The Hague, The Netherlands: Kluwer Law International, 2002. van Rooij, B., “Organization and Procedure in Environmental Law Enforcement: Sichuan in Comparative Perspective,” China Information, XVII, 2:36–64, 2003. van Rooij, B., “Implementation of Chinese Environmental Law: Regular Enforcement and Political Campaigns,” Development and Change, 37, 1:57–74, 2006.

he et al. 771

van Rooij, B., “The People vs. Pollution: Understanding Citizen Action against Pollution in China,” Journal of Contemporary China, 19, 63:55–77, 2010. van Rooij, B. and C. W. Lo, “Fragile Convergence: Understanding Variation in the Enforcement of China’s Industrial Pollution Law,” Law and Policy, 32:14–37, 2010. Wang, H., Pollution Charges, Community Pressure, and Abatement Cost of Industrial Pollution in China. Washington, DC: World Bank, Policy Research Working Paper No. 2337, 2000. Wang, H., “Pollution Regulation and Abatement Efforts: Evidence from China,” Ecological Economics, 41, 85–94, 2002. Wang, H. and M. Chen, How the Chinese System of Charges and Subsidies Affects Pollution Control Efforts by China’s Top Industrial Polluters. Washington, DC: World Bank, Policy Research Working Paper No. 2198, 1999. Wang, H. and Y. Jin, Industrial Ownership and Environmental Performance: Evidence from China. Washington, DC: World Bank, Policy Research Working Paper No. 2936, 2002. Wang, H. and Y. Jin, “Industrial Ownership and Environmental Performance: Evidence from China,” Environmental and Resource Economics, 36, 3:255–273, 2007. Wang, H., N. Mamingi, B. Laplante, and S. Dasgupta, “Incomplete Enforcement of Pollution Regulation Bargaining Power of Chinese Factories,” Environmental and Resource Economics, 24, 3:245–262, 2003. Wang, H. and D. Wheeler, Pricing Industrial Pollution in China: An Econometric Analysis of the Levy System. Washington, DC: World Bank, Policy Research Department Working Paper No. 1644, September 1996. Wang, H. and D. Wheeler, Endogenous Enforcement and Effectiveness of China’s Pollution Levy System. Washington, DC: Policy Research Working Paper No. 2336,World Bank, 2000. Wang, H. and D. Wheeler, “Equilibrium Pollution and Economic Development in China,” Environment and Development Economics, 8, 3:451–466, 2003. Wang, H. and D. Wheeler, “Financial Incentives and Endogenous Enforcement in China’s Pollution Levy System,” Journal of Environmental Economics and Management, 49, 1:174-196, 2005. Wong, Christine P. W., Fiscal Management and Economic Reform in the People’s Republic of China. Oxford, UK and New York, NY: Asian Development Bank and Oxford University Press, 1995. Wong, Christine P. W., “Rebuilding Government for the 21st Century: Can China Incrementally Reform the Public Sector?,” The China Quarterly, 200:929–952, 2009. Zhan, X., C. W. Lo, and S. Tang, Contextual Changes and Environmental Policy Implementation in China: A Longitudinal Study of Street Level Bureaucrats in Guangzhou. Los Angeles, CA: University of Southern California, School of Policy, Planning and Development, Environment and Sustainability, WP-October, 2009.