Judicializing Environmental Politics? China's Procurator-led Public Interest Litigation against the Government

Abstract Scholars consider deficient local accountability mechanisms a key shortcoming of China's response to environmental issues. Through empirical analysis of the new procurator-led public interest litigation (PIL) system, this study examines whether – and to what extent – this shortcoming can be remedied by empowering the juridical institutions. It concludes that thanks to the procuratorates’ political insider status, relative autonomy from local politics and extensive resources, procurators have generally found ways to maintain a delicate balance between holding executive agencies environmentally accountable and managing local governments’ resistance to the PIL system. However, reliance on top-down political support may ultimately hinder the expansion and stability of the procuratorial PIL system.


Introduction
Many see China's insufficient local implementation as a key problem of its environmental governance. 1 In recent years, the people's procuratoratesonce low-profile prosecutorial institutionshave rapidly become key actors in tackling this problem, thanks to their newly acquired power of public interest litigation (PIL). After a two-year pilot period, the National People's Congress formalized the PIL system by amending both the Civil Procedure Law and the Administrative Procedure Law in 2017. The authorizations include the capacity to investigate and sue both private parties and government agencies for violations against national and public interest. 2 Notably, both laws 1 Gilley 2012;Seligsohn, Liu and Zhang 2018;Eaton and Kostka 2014;Xiaoliang Li et al. 2019. 2 Civil Procedure Law of the PRC, Article 55, available at: http://www.npc.gov.cn/zgrdw/englishnpc/Law/2007-12/12/con-tent_1383880.htm; Administrative Procedure Law of the PRC, Article 25, available at: http://www.npc.gov.cn/zgrdw/englishnpc/Law/2007-12/12/content_1383912.htm. specifically list the "protection of the ecological environment and resources" first among PIL's primary aims. 3 Given their extensive geographic presence and vast state resources, procurators soon became a dominant force in legal action against environmental violations, particularly through PIL against the government. According to official records, in 2020 alone, the procuratorates officially registered and investigated 83,744 environmental PIL cases, of which 90.4 per cent were against state agencies. The report also indicates that these casesthrough either pre-litigation settlement or court judgmentresulted in about 380 million yuan (US$52 million) in damages, 3.64 billion yuan (US$502 million) in restoration fees, the closure or "rectification" of 7,702 companies and the restoration of 150 square kilometres of natural land. 4 In addition, court data show that of all environmental PIL cases concluded in 2020, 97.1 per cent were brought by procuratorates and only 2.9 per cent by NGOs. 5 The Supreme People's Court (SPC) thus announced that "the procuratorates have become the 'absolute main force' of environmental public interest litigation, which highlights the Chinese characteristics of the public interest litigation system." 6 Procurator-led PIL thus represents a possible way to hold China's vast local party-state environmentally accountable. If proved effective, procuratorial PIL against the government offers a potential institutionalized solution to one of the major deficiencies of the country's environmental governance. Through extensive field research, this study aims to explore the complex relationship between the procuratorates and the local party-state and to examine whetherand to what extentthe new PIL power to sue the government can help keep its environmental behaviours in check.

Accountability Deficiency and China's Quest for Centralized Environmental Governance
Scholars have identified the inability to hold local officials accountable as a key cause of China's environmental problems. 7 For instance, Gilley argues that implementation of central environmental policies in China suffers from policy incoherence and local pushbacks against central directives, concluding that the country's approach to environmentalism "is more effective in producing policy outputs than outcomes." 8 Similarly, Eaton and Kostka's empirical work in the early 2010s discovered that China's local officials "tend to select cheap and quick approaches to environmental policy implementation." 9 The implementation gap has been further exacerbated by the lack of accountability mechanisms other than top-down monitoring, which is often dysfunctional. Seligsohn, Liu and Zhang have found that increased local government transparency does not lead to better environmental outcomes, as the public lacks the means, such as competitive elections, to hold local governments accountable. 10 China offers a prime opportunity to examine whether these problems can be effectively mitigated, as it has recently launched wide-ranging institutional reforms to enhance environmental accountability at the local level. 11 One relatively understudied aspect of this effort is the partial judicialization of local environmental politics. Judicialization refers to "the expansion of the province of courts and judges in determining public policy outcomes" through means such as administrative review and rights jurisprudence. 12 Scholars point out that this process has been observed since the late 20th century in both established democracies like the United States and post-authoritarian European and South American states and often results from increased rights consciousness and global expansion of liberal constitutionalism. 13 However, recent studies indicate that authoritarian or hybrid regimes may also advance judicialization for various ends, including promoting trade and investment, controlling the bureaucracy and resolving the principal-agent problems between central and local governments. 14 China has followed this trend since it started establishing specialized administrative law tribunals to help rein in its vast and increasingly autonomous local party-state in 1988. 15 Xi Jinping's "lawbased governance" ( yifa zhiguo 依法治国) reforms are the latest examples of the state creating more local accountabilityparticularly in the area of environmental protectionby empowering judicial institutions. These include a series of high-profile legislative and institutional changes that have brought new actors into the field of environmental governance, such as the establishment of special environmental tribunals, NGOs' new legal standing to bring environmental litigation andperhaps most importantlythe formation of the procuratorial PIL system. 16 Promoting government accountability through judicialization, however, can present major challenges outside consolidated democracies. Moustafa finds that authoritarian governments may severely limit or even reverse the judicialization process due to concern that civil society actors may use the courts as a venue for challenging authoritarian rule. 17 Similarly, scholars argue that China's "turn against law" in the late 2000s was largely driven by the state's concern that courts would become a major source of social contention. 18 Another problem is local self-preservation: without the robust media and political oversight typical in established democratic systems, judges in administrative lawsuits are often under political pressure to resolve cases in favour of local agencies. 19 In the Chinese context, this problem is particularly prominent when the courts fail to secure endorsement from local Party leadership or a defendant is a powerful agency with strong political connections. 20 Procuratorial PIL has the potential to mitigate these problems. A regime may be wary of opening the judicial process to societal actorssuch as NGOs and rights lawyersas their use of the courts may pose "a credible threat to the regime's tools for maintaining control." 21 By contrast, even with broad power to sue government agencies, state organs like the procuratorates have little incentive to challenge the regime, making procuratorial PIL a politically "safe" form of judicial empowerment and reducing the likelihood that the government will object to judicialization on political grounds. At the same time, procuratorates' involvement in administrative lawsuits may serve to counterbalance local self-preservation. Unlike private plaintiffs, who usually have little political clout, procuratorates are major state organs that traditionally have close relationships with the courts. This makes it less likely that judges will disproportionally favour defendants in procuratorial PIL against local agencies. As juridical organs, the procuratorates are also less intertwined with the local party-state than administrative agencies and thus potentially more resistant to government interventions.
After PIL was formalized in 2017, China's highest judicial authorities promulgated a series of rules in an attempt to provide local branches with guidance and authorization to hold local agencies 12 Hirschl 2011, 256. 13 Ferejohn 2002Hirschl 2008;Sieder, Schjolden andAngell 2016. 14 Moustafa 2014;Ginsburg and Moustafa 2008. 15 Peerenboom 2002. 16 Van Rooij, Stern and Fürst 2016. 17 Moustafa 2007. 18 Minzner 2011Liebman 2014;Wang 2020. 19 Solomon 2007. 20 He 2013. 21 Moustafa 2003 accountable. In 2018, the Supreme People's Procuratorate (SPP) and the SPC jointly issued the "Interpretation on Several Issues Concerning the Application of Law for Cases Regarding Procuratorial Public Interest Litigation," which lays out the basic procedural rules for PIL. Among other things, the interpretation specifies that a procuratorate should issue a procuratorial recommendation when it finds that a government agency has acted illegally or failed to perform its duty with regard to eco-environmental protection. If an agency fails to comply with the recommendation within two months, the procuratorate can bring an administrative lawsuit against it, and the court can then rule to void, reverse or change the agency's illegal actions as well as compel the agency to discharge its legal duties. In 2021, the SPP issued the "Rules for Handling Public Interest Litigation Cases," which address some problems associated with administrative PIL. For instance, the rules stipulate that "if an administrative agency … refuses or jeopardizes a procuratorate's investigation, the procuratorate can report to the people's congress on the same level, report to the disciplinary agency on the same level, or report to the agency's supervising authority through the higher procuratorate." 22 However, without empirical inquiry into the inner workings of procuratorial PIL, it is difficult to assess the system's efficacy at holding local agencies accountable.
Indeed, scholars strenuously disagree on whether procuratorial PIL can effectively address China's problems with implementing environmental governance. Comparing China's procurator-led civil PIL system to its Brazilian counterpart, Shi and Van Rooij argue that the Chinese system is plagued by institutional dependence on the local party-state, inadequate expertise in specialized environmental cases and a lack of "major political change in any way comparable to the democratization processes following military rule in Brazil." 23 Claiming that these shortcomings cause local procurators to take only easy cases with minor environmental impact, they predict that Chinese procurators "are unlikely to become a strong regulatory force." 24 Based on case data from 2003 to 2018 and interviews with twelve procurators and judges, Ding and Xiao also find that poor competence and institutional autonomy have led local procuratorates to prefer low-hanging fruit in civil PIL. 25 Liu similarly points out that procuratorates brought less administrative PIL to court in 2019 than in 2017, attributing this to "resistance from external factors," such as government agencies' unwillingness to be sued. 26 In contrast, Zhai and Chang argue that its financial resources, professional staff and ability to collect evidence from government agencies "make the procuratorate the most powerful actor in public-interest litigation, which helps … when powerful enterprises or administrative agencies act as the defendant." 27 This optimism also finds some support from Chinese empirical studies: Examining data from 287 prefecture-level cities from 2010 to 2018, Liu and Fan discovered that procuratorial PIL has effectively reduced emissions of industrial wastewater and sulphur dioxide. 28 More specifically, Shen and Xing argue that during the experimental stage of administrative PIL (2015-2017), the procuratorates had successfully compelled agencies to rectify their actions through the use of procuratorial recommendations. 29 Through one-on-one, in-depth interviews with 31 PIL procurators, this study offers the first comprehensive examination of the Chinese procurators' role in holding local governments environmentally accountable. To mitigate potential self-serving bias, findings were corroborated by interviews with employees from five major national environmental NGOs that had worked extensively with the procuratorates on PIL-related issues. Unlike the studies by Shi  Rooij and Ding and Xiao, which primarily examined civil PIL against private parties, this study focuses on administrative PIL against local government agencies. These administrative cases not only represent around 90 per cent of PIL procurators' registered cases and actual workloads but also are directly relevant to China's problems with implementing environmental policies. This article also provides a timely update to existing empirical studies on the subject; as the procuratorial PIL system was not formalized until mid-2017, earlier data likely do not reflect the system's operation beyond the initial trial-and-error period. This additional time is particularly important, as it may have allowed the procuratorates to develop expertise in environmental litigation, thereby mitigating the competence concerns raised by their critics.
The study finds that the procuratorates generally maintain a delicate balance between increasing the environmental accountability of local agencies and managing the agencies' considerable resistance to the PIL system. Building upon the hierarchical procuratorial apparatus, PIL teams operate under a highly centralized structure, with provincial and intermediate procuratorates holding immense control over local procurators. They also benefit from endorsements from high-level central and provincial leadership, which provide crucial legislative and political leverage against the local party-state. Consequently, the procuratorates can pressure local government agencies to implement environmental laws and regulations through the threat of litigation, even though local procurators often rely on collaboration with the local party-state during PIL investigations. However, dependence on high-level support may constrain expansion of the PIL system and render it susceptible to political vicissitudes. The study thus contributes to the debate surrounding the efficacy of China's approach towards environmentalism, especially regarding its traditionally insufficient accountability mechanisms. It also adds to the growing literature on judicialization of politics in non-Western countries.

Methodology
This study is based on semi-structured interviews that the authors conducted in 2021. The fieldwork included interviews with 31 procurators from five province-level units: affluent Beijing, moderately prosperous Hubei and Shandong, and economically under-developed Jilin and Guizhou. Interviewees included seven procurators from two provincial procuratorates, 10 from three intermediate procuratorates, and 14 from seven basic procuratorates. All interviewees worked on PIL in some capacity, such as through frontline prosecution or overseeing cases. The relatively low number of interviewees per procuratorate was due to the small size of PIL teams, particularly at the basic procuratorate level.
The interview questions focused on how the procuratorates perform their newly acquired PIL function. They also addressed other relevant issues, such as relationships with local governments. The interviewees were asked to focus their answers on the period after the PIL system was formalized in 2017. The following are some typical questions: (1) Who are the people serving on your public interest litigation team?
(3) How do you find environmental public interest cases? (4) How do you decide whether to formally investigate a case? (5) What are the main means of investigation? How well do government agencies cooperate? (6) How does the procuratorial recommendation process work? (7) How do you decide whether to bring a public interest lawsuit to court? (8) How do you evaluate an agency's compliance with a procuratorial recommendation or court order? (9) How do you balance the procuratorate's relationships with agencies that you sue or issue procuratorial recommendations to?
These questions comprised a general interview guide for the authors. If an interviewee appeared particularly knowledgeable about or interested in a specific topic, however, the authors would encourage deeper conversation. The authors also asked follow-up questions not included in the list above. Interviews typically lasted one to two hours. To mitigate potential self-serving bias, the authors interviewed 10 employees of five major national NGOs specializing in environmental litigation to corroborate the procurators' accounts. Having collaborated with the procuratorates on environmental PIL cases, many employees from these organizations had substantial knowledge on the subject. Interviewees were asked to comment on the efficacy of procuratorial PIL, particularly at holding local governments environmentally accountable.
To encourage candid discussion, interviewees were promised that their names, as well as the names of their cities and counties, would not be included in any publications. The interviewees' have thus been designated by three letters and two digits (e.g. PBJ01) according to the following method. The first letter (P or N) denotes "procurator" or "NGO employee." The second and third letters indicate the interviewee's province (BJ=Beijing, HB=Hubei, SD=Shandong, JL=Jilin, GZ=Guizhou).

The Procuratorial PIL System: A Centralized Structure
Historically, the people's procuratorates have been centralized and hierarchical institutions. According to the Organic Law of the People's Procuratorate of the PRC, higher procuratorates "lead" the work of lower procuratorates by giving direct orders and revoking policies and decisions at will. 30 Consequently, Chinese procurators generally follow a clear chain of command from the SPP to the grassroots, 31 and the new PIL system has largely adopted this top-down structure, with higher procuratorates frequently issuing both broad policies and specific orders to local subordinates.
The four-level hierarchy Like the rest of the people's procuratorate apparatus, the new PIL system operates under a four-level structure that gives higher procuratorates substantial control over local subordinatesan arrangement critical to the procuratorates' ability to resist local intervention. At the top are the PIL divisions in the SPP and the provincial procuratorates, which rarely bring PIL, focusing instead on providing policy guidance to the intermediate and basic procuratorates. For example, the provincial procuratorates often issue internal policy guidance to their subordinates and use performance evaluations and other control mechanisms to induce compliance, an issue discussed in more detail in the next section.
Most case-related work is completed by PIL teams at the intermediate and basic levels. Intermediate procuratorates, whose jurisdiction usually covers a sub-provincial region or large metropolis, serve as bridges between the policy-makers in provincial procuratorates and the litigators in basic procuratorates. For instance, they are generally responsible for evaluating the basic procuratorates' performance using criteria set by the provincial procuratorates. They also handle a relatively small number of administrative PIL cases against agencies above the county/district level. At the bottom of the hierarchy are the basic procuratorates, which bear the brunt of PIL work, including all administrative cases against grassroots agencies. This is because most of the state's day-to-day administrative functionsincluding environmental protectionare performed by agencies at or below the county/district level. Besides the abovementioned division of work, the hierarchical nature of the PIL system is evident from PIL team size at each level. PIL divisions in the provincial and intermediate procuratorates average around 10 people, and usually more than half are procurators and the rest are clerks. 32 In contrast, most basic procuratorates lack dedicated PIL divisions, 33 averaging only one or two procurators dedicated to PIL work and one or two clerks or judicial police officers to assist. The structure signals a clear intention to establish a strong chain of command that can effectively turn the SPP's broad mandates into operable and localized policies on the ground.
Tools of control: performance evaluations, special projects and case instruction Several tools are available to the higher procuratorates to rein in the PIL teams in their jurisdictions. First and foremost is the performance evaluation system. A provincial procuratorate sets the general criteria for the entire province, which form the basis for annual evaluations that pit local procuratorates against one another. 34 As these evaluations guide promotion of local procuratorate leaders, frontline procurators are generally under significant pressure to follow the provincial procuratorate's policy priorities. 35 For example, in 2019 the SPP found that the procuratorates were bringing too many "easy" PIL cases against criminally convicted individuals and called on them to bring more cases against government entities. 36 In response, the Jilin people's procuratorate increased the "point value" of administrative PIL, making it double that of PIL against convicted persons. 37 The change proved a significant incentive, as administrative PIL cases represented more than half of Jilin's total 2020 PIL count, compared to only around 20 per cent in many other provinces. 38 The provincial procuratorates also frequently announce "special projects" (zhuanxiang 专项), which are programs designed to encourage lower procuratorates to bring certain types of PIL. Examples include projects on managing medical waste and reducing plastic bag use. 39 To acquire high-level recognition and good publicity, provincial procuratorates often design such projects around either the latest Party or SPP policies or hot-button topics in the media. 40 Some intermediate procuratorates issue "sub-projects" based on these provincial projects to tailor the schemes to local conditions. 41 Local procurators generally value the extra rewards offered by special projects and thus dedicate significant time to relevant cases.
By far the most direct means of top-down control is the case approval system. Under this system, a lower procuratorate must seek approval from a higher one before making certain moves during a PIL case, such as issuing a procuratorial recommendation or bringing a lawsuit, though the exact rules differ from province to province. Through this system, higher procuratorates can directly control PIL-related political risk and reduce some principal-agent problems, such as a lower-level procuratorate conspiring with a defendant to withdraw a case. 42 As discussed later, this power becomes particularly important when a higher-level procuratorate attempts to shield PIL cases from local government interference.
In sum, like the rest of the people's procuratorate system, the four-level PIL structure is hierarchical and centralized. This is particularly true because provincial and intermediate procuratorates not only set evaluation criteria for their local subordinates but also frequently give direct instructions for individual cases. As discussed in the next sections, such a centralized structure inevitably comes into tension with the local nature of PIL cases.
Environmental PIL Cases against the Government: Navigating Local Dynamics Like many other legal issues in China, administrative PIL cases are deeply embedded in the political dynamics of the local party-state. Studies have found that powerful local actors often try to influence the decisions of the local judiciary. 43 It is therefore reasonable to assume that procuratorial PIL will face similar resistance from localities. 44 The following section examines the complex relationships between procuratorates and local government institutions to determine how much pressure the latter can exert on PIL procurators.
Pre-litigation collaboration with the government Collaborations with local government agencies are crucial during the pre-litigation stage of administrative PIL, which mainly involves finding case leads, identifying agencies to investigate, and conducting investigations. Interestingly, many administrative PIL cases brought by procuratorates arise through formal or informal agreements with government entities. Some procuratorates manage to establish long-term information sharing agreements with specific agencies, which help procurators identify flaws in the agencies' official actions. One example is Hubei, where 16 government agencies have agreed to upload their internal administrative information to an online platform accessible by the procuratorates, 45 allowing procurators to browse administrative penalty decisions, including those made by environment-related agencies against polluters. When procurators suspect penalties are insufficient or otherwise ineffective, they send teams to check whether the pollution has ceased, which often reveals continued pollution and leads to PIL inquiries against the agencies. 46 Even more ironically, local government institutions sometimes ask to be investigated by the procuratorates, hoping to leverage the latter's legal authority to advance their own policy goals. For example, a district environmental protection agency in Hubei once asked the procuratorate to launch a PIL investigation against its inaction in water and soil conservation. The agency's inaction was due to a lack of political clout to compel a large university within its jurisdictionwhich outranked the agency in the administrative hierarchyto pay for water and soil conservation on its campus. Unwilling to confront the university on its own, the agency asked the procuratorate to issue a formal recommendation against its "failure" to compel the university. The agency then forwarded the recommendation to the university, explaining that it was "required by legal authority" to punish the latter for noncompliance. 47 Collaboration with the government is also vital to identifying agencies to investigate. After finding a case lead, the first step is usually determining whom the case should be against, which in theory should be the government entity responsible for the administrative action or inaction in question. Although identifying the agency behind a government act is straightforward, it is often difficult to pin down responsibility for inaction (which is the subject of most administrative PIL). This is largely because agencies' functions overlap and the laws that allocate responsibility are often vague. For example, in a case involving pollution caused by vehicle spray-painting services, the local procuratorate found it difficult to determine whether such businesses should be regulated 43  by environmental protection bureaus (EPBs) or transport management bureaus. 48 Under such circumstances, the relevant government entities often hold a roundtable discussion to determine the proper target of the PIL. 49 This approach, however, requires voluntary participation from these agencies, which are usually under no legal obligation to act, especially since they have not yet been formally investigated. 50 Perhaps most importantly, the procuratorates rely heavily on local agencies to collect key evidence. Many PIL casesespecially environmental onescall for technical evidence that can only be collected through professional testing that is beyond the expertise of most procuratorates. Because commercial testing services are scarce and expensive in many areas, relevant government agenciessuch as the EPBs and water authoritiesare usually the local procuratorates' primary sources of technical evidence. The agencies' assistance on such matters, however, is largely discretionary since no national laws compel them to collaborate with the procuratorates. This gives these government institutions considerable leverage against the procuratorates, especially regarding PIL.
Collaborations with agencies like EPBs become especially necessary when a procuratorate must investigate private parties. Many interviewed procurators noted that they lack legal authority to compel private parties to comply with their investigation requests during civil and administrative PIL. Consequently, they often need help from agencies with the power to inspect private parties. For example, a district procuratorate in Jilin often asks officials from the local EPB to accompany procurators during administrative PIL investigations, so that companies cannot resist the procurators' requests to test their production sites for air pollutants. 51 In addition, the loss of their anti-corruption bureaus has increased procuratorates' reliance on voluntary collaborations with local agencies. Procuratorates previously held considerable power over local government institutions due to their ability to launch anti-corruption investigations, a power they often used to coerce agencies to comply with their PIL investigation requests. 52 However, procuratorates' political status has fallen greatly since losing this anti-corruption function in the 2018 reforms that included the founding of the National Supervisory Commission (NSC), as local agencies no longer bow to their demands from fear. 53 One procurator complained that some government agencieswith which she previously had good working relationsbegan to give her the cold shoulder immediately after the procuratorate lost its anti-corruption branch. 54 As a result, the cooperation of such agencies increasingly hinges upon goodwill.
This situation now drives many procuratorates to seek amicable and mutually beneficial relationships with local agencies. For example, in 2021, the Hubei people's procuratorate instructed the province's procuratorates to "form win-win relationships" with these agencies. 55 A key part of the campaign was encouraging procurators to negotiate with relevant agencies to reach settlements on rectifying the latter's wrongdoings before issuing procuratorial recommendations or bringing lawsuits. According to an interviewed procurator from the Hubei people's procuratorate, the policy's objective was to facilitate collaborative communications between the procuratorates and the executive branch, which would benefit PIL practice. 56 Another example is the liaison system developed by the Jilin procuratorates, which established hotlines between procurators and key officials in government entities central to PIL work. 57 However, such win-win relationships inevitably hit their limit when it comes to administrative litigation in court.

The cost of litigation
The true test of the relationship between a procuratorate and a local environmental agency is when the former must decide whether to sue the latter in court. Upon concluding an investigation, the procuratorate issues a recommendation to the relevant government agency that details the steps it must take to rectify its alleged wrongdoing. If the agency fails to fully comply within the designated time (usually two months), the procuratorate is theoretically obligated to bring an administrative lawsuit against it. 58 In practice, however, procuratorates have substantial discretion in setting compliance deadlines and determining compliance with their recommendations, which means they often ultimately decide whether to bring lawsuits.
Such decisions are important, as, depending on local evaluation criteria and disciplinary practices, being sued by a procuratorate can spell serious trouble for a government agency. The interviewed procurators stated that they often communicate with the local court about a case's merit before and after bringing a lawsuit, thereby ensuring the judge will support their claims against the defendant agency. 59 The procuratorates' claims in such cases are usually similar to their original procuratorial recommendations, such as requiring the defendant to re-investigate certain environmental issues or punish a particular polluter. However, the real "bite" of such litigation is not the court remedies but rather the political ramifications that come with these lawsuits. In some regions, becoming a defendant in an administrative PIL case can negatively impact an agency's performance evaluation. 60 More importantly, many local branches of the newly established NSCwhich is responsible for disciplining government officialsalso periodically collect information about the latest PIL cases filed in court. This causes anxiety among officials, who fear such investigation will reveal information that will prompt disciplinary action against their misbehaviour or inaction (bu zuowei 不作为). 61 Even when these serious consequences are avoided, PIL can create major public embarrassment for the government, especially since today's courts often demand that the head of a defendant agency personally appear before the judge. 62 It is thus not surprising that most local agencies are eager to avoid being sued by the procuratorates.
That being said, it should be noted that the vast majority of administrative PIL cases have not been brought to court. According to data released by the SPP, less than 1 per cent of administrative PIL investigations have ended up in litigation; most have resulted in only procuratorial recommendations. 63 However, given the procuratorates' greater than 99 per cent PIL winning rate, the few litigation cases nonetheless serve as a "sword of Damocles" for potential defendants, placing substantial pressure on them to promptly comply with procuratorial recommendations. 64 One procurator explained: Most agencies did not realize they might actually get sued by us until we sued them for the first time. They thought we would only issue procuratorial recommendations, so they had no idea how serious things could get. educational experience. Most of the agencies that attended the trial became much more proactive in responding to our procuratorial recommendations. 65 Despite the few lawsuits, the potentially serious consequences of litigation strain procuratorates' relationships with local environmental agencies, whose collaboration is crucial during the prelitigation stage of PIL. Many local procurators are reluctant to bring lawsuits against these agencies, fearing that damaging these relationships will negatively impact their future investigations. 66 On this point, one Hubei basic-level procurator's experience is representative: A few years ago, we brought a lawsuit against the water authority. They came to us in private and pleaded that we do not sue them. We did not agree, which badly offended them. For quite a long while, they did not help us test for water pollutiontests we needed as evidence in other cases. This is one reason that we do not sue the environmental protection bureau, as we need its help to test for many substances. It would cost a lot of money to hire a commercial testing company to do these tests. 67 In some types of cases, a procuratorate might even face direct interventions from the top leadership of the local Party or government systems. A typical scenario is an environmental lawsuit that may hamper local economic growth. 68 For example, a Guizhou basic procuratorate has received multiple requests from the county government to halt administrative PIL against illegal land use approvals for new businesses, as officials worry such lawsuits could scare off investments. 69 Although such interventions from the top brass of the local party-state are not common, given the Party's broad power and influence over localities, they are generally difficult for local procuratorates to resist.

To Sue or Not to Sue? How Procuratorates Try to Overcome Local Pushback
Here is the key question: can procuratorates hold local government institutions accountable for their environmental actions or lack thereof despite the local resistance? Obviously, the combination of the procuratorial PIL system's centralized structure and its reliance on the local party-state creates a political dilemma for procurators trying to decide whether to sue a local agency. From the perspective of the SPP and the provincial procuratorates, the term "public interest litigation"as opposed to "supervision" or "mediation"implies that lawsuits are needed to justify this newfound power and raise awareness of it within other government institutions. 70 Indeed, as mentioned previously, the SPP and some provincial procuratorates have specifically instructed local procuratorates to bring more PIL cases against government institutions. Given the top-down nature of the procuratorial system, such requests put significant pressure on local procurators to bring more lawsuits. On the other hand, the need to collaborate with government agenciescoupled with occasional interventions from the local party-state leadershipoften give local procuratorates pause when weighing whether to litigate. As one basic-level procurator from Shandong summarized, Based on the SPP's guidance, both the provincial and municipal procuratorates have recently been demanding that we bring more administrative lawsuits. This has become a major headache for us … Much of our work requires collaboration with government agencies and suing them will make such collaboration difficult. Recognizing the problem, provincial and intermediate procuratorates sometimes try to mitigate this dilemma through direct intervention. One common method is to pressure local agencies by negotiating with their superiors. If a local EPB refuses to cooperate with a basic procuratorate during a PIL investigation, for example, the intermediate procuratorate can ask the prefectural EPB to instruct it to comply with the investigation request. 72 In some cases, a provincial procuratorate may contact province-level government institutions to increase the pressure on local agencies to collaborate. 73 Another form of intervention is case reassignment. According to SPP guidelines, a higher procuratorate can reassign a PIL case from one lower procuratorate under its jurisdiction to another or to itself "based on what is needed to handle the case." 74 Pursuant to this rule, intermediate procuratorates sometimes reassign administrative PIL cases that raise serious concerns about local intervention from one basic procuratorate to another. 75 In more significant cases, a higher procuratorate might decide to take the matter into its own hands. One procurator from an intermediate procuratorate said: [Some officials] might be really upset [about an administrative PIL case], in which case they might try to influence our local procuratorates. If there is too much pressure, we will go there and personally work on the case. Since [the local governments] have little prior relationship with us, pushing the case through becomes easier. 76 By far the most effective way for a procuratorate to overcome this dilemma is to obtain political support from the provincial Party apparatus. According to an environmental NGO lawyer with high-level government connections, such support has come at least partly from deliberate lobbying by the SPP: Initially, the local procuratorates were cautious and afraid of offending the local governments. The SPP then sent its own PIL team to some provinces to talk with the governors and provincial party secretaries, asking them to work with the local procuratorates and promise to never blame them [for suing local governments]. Only after that did the procuratorates begin to act increasingly boldly [in environmental PIL cases]. 77 This kind of endorsement can prompt the provincial congresses to pass legislation that offers local procuratorates legitimacy to deal with the local party-state. 78 For instance, after the Jilin Party committee became the first provincial Party authority to formally endorse procuratorial PIL, the Jilin people's congress soon thereafter issued the "Decision to Strengthen Procuratorial Public Interest Litigation" in 2019. 79 This local legislation requires all agencies to cooperate with PIL investigations and comply with procuratorial recommendations and encourages local governments to include procuratorate collaboration in their officials' performance evaluations. It also stipulates that all courts in the province must proactively enforce PIL judgments. 80  Perhaps more importantly, support from the provincial Party leadership boosts a procuratorate's political status, enabling it to take a firmer stance against the local party-state. Interviewed procurators in Jilin often credited the provincial Party committee's political endorsement with their high number of environmental PIL compared to other provinces. 82 One way the procuratorates leverage such endorsement is by reporting agencies that refuse to collaborate with them to the committee, thereby forcing these agencies to comply with their requests. 83 Similarly, support from provincial Party leadership helps a procuratorate resist intervention from local governments. For example, a basic-level procurator from Jilin reported that it recently disregarded the municipal government's request to suspend certain environmental PIL, choosing to nonetheless bring the lawsuits "because we have already spent a lot on the evidence." 84 Furthermore, backing from a Party committee emboldens a procuratorate to use confrontational tactics during administrative PIL, such as inviting new media to report on a case. 85 Some may even exert pressure on particularly unruly government entities by holding public hearings that include local congress members and journalists. 86 On the other hand, absent strong political support from provincial leadership, a procuratorate may find it difficult to resist pressure from local governments in certain types of administrative PIL. For example, in the previously mentioned case in which local leadership pressed a basic procuratorate to drop cases against illegal land use approvals, the Guizhou provincial procuratorate folded, instructing the local procuratorate to abandon the cases and not bring similar lawsuits in the future. 87 Such examples, which contrast starkly with the experiences of the Jilin procurators, showcase the key role of provincial leadership in shielding administrative PIL from local intervention.
However, despite the difficulty of bringing lawsuits in court, procuratorial PIL work has substantially impacted local governments' compliance with environmental laws. Across provinces, interviewed procurators reported experiencing significant changes in environmental agencies' interactions with them after the PIL system was established. Initially, many agencies treated these procurators with either contempt or extreme caution, as they were unsure how the procuratorate would wield its new power. 88 It was not until the first local PIL lawsuits were brought and the agencies became better acquainted with the procurators that the two began to work closely together. Although such collaboration has made suing the agencies more difficult, it has also made the agencies more aware of the legal risks of their actions (or inaction). The interviewed procurators generally agreed that the environmental agencies they frequently worked with had become more law-abiding and quicker to comply with their recommendationsan outcome that has ironically made their jobs harder because there are fewer causes to litigate. 89 This effect has been strengthened by the so-called "looking back" (huitou kan 回头看) mechanism, which requires procurators to periodically conduct post-PIL checks on agencies that have received procuratorial recommendations. If an agency's compliance is found to lapse, the procuratorate will immediately relaunch a full PIL investigation. 90 Even in cases of interference by local leadership, the interfering officials typically will eventually push for compliance with procuratorial recommendations if the procurator promises not to sue.

Discussion and Conclusion
As demonstrated above, the procuratorial PIL system has generally found ways to balance its quest to hold local agencies environmentally accountable and its reliance on government collaboration during the pre-litigation stage. The foundation for resisting local intervention is embedded in the PIL system's centralized structure. The new PIL teams have largely inherited the hierarchical structure of the procuratorial apparatus. More specifically, higher procuratorates can assert a great deal of control over local PIL teams through various mechanisms, including performance evaluations, special projects and the case approval system. This centralized structure is most seriously tested when it encounters environmental PIL cases against local governments. Procurators need help from local agencies with many pre-litigation tasksfrom identifying agencies that warrant investigation to testing pollutant samples. In particular, procuratorates rely heavily on local environmental protection agencies, such as the EPBs and water authorities, to provide substance-testing expertise and the legal authority to investigate private companies. Such reliance has been further exacerbated by the procuratorates' loss of their anti-corruption function in 2018, which has deprived procurators of an important channel for compelling local agencies. This has created a dilemma for local procurators, who are under top-down pressure to investigate and sue the same agencies they frequently work with for violations and inaction related to the environment.
In practice, procuratorates are generally able to mitigatethough not completely overcomethis dilemma thanks to a combination of internal top-down intervention and external political endorsement. When a PIL case is at risk of local interference, a higher-level procuratorate can reassign the case to another local procuratorate or to itself, thus decreasing the local government's ability to influence the litigation process. In some regions, high-level political support, such as an official endorsement by the provincial Party leadership, may also alleviate pressure from the local party-state. Such endorsement not only nudges the provincial congress to pass pro-PIL legislation but also raises a procuratorate's political status, allowing it to take a more confrontational approach towards local agencies on environmental issues. Though the vast majority of administrative PIL investigations only end up in procuratorial recommendations rather than lawsuits, the mere threat of litigationswhich may lower performance evaluations or even prompt disciplinary actionsstill impose substantial pressure on local agencies to promptly implement the procuratorates' recommendations.
The procuratorial PIL system thus offers an example of how a state can mitigate its implementation problems through the process of judicialization. The hierarchical and professional nature of the juridical institutions gives them a unique advantage againstthough not immunity fromillegitimate local interference. 92 Courts and procuratorates are generally less embedded in the local power complex than most executive agencies because they are legally and administratively obligated to follow national laws as interpreted by their superiors. This is especially true for the procuratorial PIL system, as its hierarchical structure enables the higher procuratorates to practically micromanage their local subordinates, making it difficult for local actors to leverage political or personal connections to avoid environmental accountability. This structural advantage is particularly pronounced when coupled with high-level political backing, such as national legislation and endorsements by provincial leadership. Such support emboldens the juridical institutions when dealing with local governments and can transform legal outcomes into political consequences for violators.
The role of judicialization in environmental governance is especially prominent in regimes that lack non-government means for holding local governments accountable. In established democracies, robust public participationlargely through elections and civil societyprovides substantial checks on officials' environmental behaviours. These types of accountability mechanisms, however, are often absent in more closed polities. Although such states sometimes tolerate or even encourage 92 He 2020. NGOs' involvement in local environmental governance, such participation is usually cautious, piecemeal and restricted in terms of formality and subject matter. 93 Indeed, that NGOs are authorized by law to bring PIL against private parties but not government agencies demonstrates the limits of civil society to enhance the environmental accountability of the local party-state. 94 Juridical institutions, such as courts and procuratorates, are thus among the few actors that possess some degree of autonomy from local governments as well as the capacity to hold them accountable.
This article also has broader implications for studying judicialization beyond consolidated democracies. Scholars point out that a major obstacle to judicialization is some regimes' concern that civil society actors might use the courts to challenge their rule. 95 The findings of this study suggest that a government might have it both ways by permitting state-controlled institutionsrather than societal actorsto bring lawsuits the state deems politically risky. This approach allows the state to realize the functional potential of judicialization while minimizing its associated political risks. Indeed, that some provincial party leaderships have actively endorsed procuratorial PIL illustrates the regime's comfort with utilizing such state-centred forms of judicialization. In addition, involving state institutions like the procuratorates brings human, financial and political resources unavailable to civil society groups, such as legal professionals in every county and connections with judges and provincial leaderships. 96 These advantages vastly accelerate the judicialization process in certain policy areas, in terms of both case volume and geographic reach. They may also enable the state to mitigate local self-preservation by counterbalancing local actors' influence over the courts.
That being said, as its influence continues to grow, procuratorial administrative PIL will likely be increasingly constrained by the procuratorates' status as state organs. As indicated by several interviewed NGO employees, though the procuratorates have been relatively successful at leveraging their resources and connections to enhance the environmental accountability of local agencies, procurators remain cautious and selective in bringing administrative lawsuits due to their political-insider status. The interviewees thus expressed concern that local procuratorates would hesitate to tackle fundamental problems in China's environmental governance, as it could require challenging powerful and high-level state institutions. 97 This dilemma is likely to become particularly pronounced given the SPP's continuous quest for institutional legitimacy and political relevance, which has increased top-down demand for high-impact administrative PIL. This ambition will inevitably hit its political limit when it clashes with key interests of the local or provincial party-state, whose support remains crucial to the functioning of the procuratorates.
Moreover, the procuratorial PIL system's heavy dependence on high-level political endorsement sheds doubt on both the universality of the judicialization approach and its sustainability in the Chinese scenario. The strength and availability of political backing for the judicialization of environmental politics hinges upon the policy priorities and power dynamics of the national leadership. For example, a similar process during the Hu Jintao era, which saw the courts playing an increasing role through pollution compensation lawsuits, yielded only marginal and inconsistent results. 98 This was due in no small part to the political fragmentation that prevented the administration from effectively reining in the local party-state and offering sufficient support to the courtseven if it intended to. 99 In contrast, the current Chinese administration has not only demonstrated considerable commitment to both protecting the environment and strengthening legal institutions but also proved highly effective in consolidating power within both the central government and centre-local 93  relations. 100 This combination of political priority and efficacy has been critical in developing political backing for procuratoratesand, arguably, the massive PIL apparatus itself. It is therefore questionable whetheror to what extentsuch pro-environment judicialization could be replicated by other polities. Indeed, considering the vicissitudes of factional politics and centre-local relations in the past decades, it remains unclear whether the relative effectiveness of the Chinese procuratorial PIL system will be sustained in the long term.