Participatory budgeting and the party: Generating ‘citizens orderly participation’ through party-building in Shanghai

Abstract This article provides a qualitative examination of two cases of Participatory Budgeting (PB) in Shanghai – a long-running PB initiative in Minhang District, organised in cooperation with the District People’s Congress, and a one-off project in Yangjing Sub-district, Pudong, in 2017, jointly organised by a community foundation and residents’ committee. The article seeks to interrogate the relationship between the Party, state and society at the sub-municipal level through one ‘state-facing’ PB initiative and one ‘society-facing’ PB initiative. We reveal how PB is deeply embedded in Party structures and networks, formally in the case of Minhang and informally in the case of Yangjing. Our research contributes to three debates on participatory governance in urban China. Firstly, contrary to the existing literature, PB neither primarily ‘emancipates’ citizens nor off-loads budgetary decisions onto them; instead, PB contributes towards party-building and citizens’ orderly participation, thereby strengthening overall Party leadership. Secondly, we challenge the widely-used term ‘party-state’, instead separating out these three entities and showing how they serve distinct roles in grassroots governance innovations such as PB. Thirdly, we show how participatory mechanisms developed in one political and cultural context can have vastly differing effects when employed in another.

by a local social organisation in cooperation with a Residents' Committee (ju weihui 居 委会, hereafter, RC) in Yangjing Sub-district, Pudong, in 2017. 2 These two cases enable us to explore the relationship of the Party to state institutions, in the case of Minhang, and to formally non-governmental organizations (NGOs), in the case of Yanjing Subdistrict. Our research demonstrates that, although the relationship between the PB initiative and the local Party is different in both cases, with the first exhibiting direct supervision by the Party, and the second demonstrating a more subtle interweaving into Party structures, the CPC is deeply integrated into these grassroots participatory mechanisms. This finding adds a new dimension to existing research on participatory governance in China, which has not yet fully mapped the dynamics between PB and the Party. It also shows how mechanisms such as PB, utilised in some contexts as forms of radical direct democracy, can be used to reinforce the political status quo in others.
Globally, political party support for PB and other deliberative initiatives, such as Deliberative Polling (hereafter, DP), 3 is widespread, and can be divided into three types. Firstly, and most frequently, localities controlled by political parties promoting a democratising or participatory agenda may organise budgetary deliberations at the municipal or district level, either regularly or on an ad hoc basis. In Brazil, local governments controlled by the left-wing Workers' Party were the key initiators of the now famous PB experiment. 4 In Germany, the Left Party instituted a PB project in Berlin-Lichtenburg. 5 In Kazakhstan, following an announcement by President Kassym-Jomart Tokayev of the ruling Nur Otan party that government should enhance its 'inclusivity', district governments have implemented PB projects. 6 Secondly, in several countries, budgetary deliberations have been enshrined in national law by ruling party-dominated legislatures: in Russia, in 2020 the ruling United Russia party, introduced two laws aiming to institutionalize citizen participation in local budgetary decisions. 7 In Mongolia, the ruling Mongolian Peoples' Party introduced a law requiring constitutional changes to be approved by DP. 8 Thirdly, national governments may organise one-off deliberations on salient social or economic issues: in Japan, DP was used by the ruling Democratic Party of Japan to help them understand public attitudes towards nuclear energy following the 2011 Fukushima nuclear accident. 9 Thus, in democracies and non-democracies, political party support for deliberative and participatory initiatives is important, not least because it is those same parties that allow or deny public access to the government departments under their control. However, our research suggests that, in China's one-party system, CPC involvement in PB goes beyond both local endorsement and national regulation, towards a full interweaving of grassroots Party structures into PB initiatives. This finding corresponds to Zheng's conception of the CPC as 'organisational emperor', a term that highlights the Party's continued domination over state and society. 10 Few studies have hitherto traced CPC involvement in PB. Some mention the CPC in passing: for instance, He's study on participatory governance in Zeguo, Zhejiang, describes how a DP event was supported by a local CPC official who had attended an international conference on deliberative democracy; however, his research focusses on the relationship between DP and PB, rather than elaborating the Party's role in delivering deliberative governance. 11 Yet in order to understand precisely why the Chinese government has so strongly advocated participatory mechanisms, which, as we discuss below, seek to involve citizens more extensively in local governance, we must understand the role of the CPC in their execution. We suggest that CPC presence in participatory institutions is precisely what makes citizen participation in China stable (or, to use official Chinese terminology, 'orderly'). Thus, we must distinguish between the structures of government and Party at the local level, and consider how participatory mechanisms interact with them both. In doing so, we take up Snape and Wang's important call for research that 'not only brings the Party in, but that examines it in relation to state and society.' 12 Questions about the ideal relationship between Party, state and society quickly lead to questions concerning constitutional reform. 13 In the socialist tradition, citizen engagement and local self-governance form two central principles. 14 Article 27 of the Chinese Constitution states, 'All state organs and state employees must rely on the support of the people, stay engaged with them, listen to their opinions and suggestions, accept their oversight, and work hard to serve them'. 15 In practice, however, the government's top-down structure and lack of incentives for officials to engage citizens often leads to officials' disinclination to implement participatory governance that goes beyond simple complaint-making. 16 In contrast, participatory activities under CPC auspices aim to strengthen support for the Party. 17 And at the street level, RCs often depend on citizens' participation in to implement policies set from above. 18 Our exploration enables us to examine how these three forces come together on the ground through PB: we suggest that the close relationship between PB and the Party constitutes evidence for how bottom-up governance mechanisms can be engineered to support the broader, top-down project of Party-building because it enables Party representatives to guide citizens towards making decisions and adopting behaviour aligned with the Party agenda.
In the last 40 years, state bureaucracy in China has undergone a profound transformation, away from the so-called 'iron rice-bowl', which saw benefits delivered directly to citizens through their workplace, towards the incorporation of market principles into public administration. This shift demanded a radical reconfiguration of the relationship between state, Party and society. On the first, government bureaucracy in China has adopted many of the norms of New Public Management (NPM), characterised as a shift 'from government to governance', 19 involving the scaling back, decentralisation, and outsourcing of services formerly provided by government agencies. 20 Scholars highlight the flexibility of local administrations, 21 their embrace of public-private partnerships and service-providing non-governmental organisations, 22 and their expansion of opportunities for citizens to engage in policy-making. 23 A growing body of research now characterises Chinese bureaucracy as a type of regulatory state, which manages and oversees the provision of public goods and services, rather than delivering them directly. 24 Transformations in self-portrayals of the Party from 'vanguard of the working class' to 'representative of the people' have accompanied this shift. 25 At the Sixteenth Party Congress in 2002, Jiang Zemin introduced the 'Three Represents', which stated that the CPC must 'represent the development trend of China's advanced productive forces, the orientation of China's advanced culture, and the fundamental interests of the overwhelming majority of the people', expanded Party membership to include China's growing business and managerial class, thereby extending Party influence into private corporations. Clearly, this signified an attempt to develop a more inclusive conception of the Party, relevant to the rapidly growing middle class. 26 Over the following years, the Hu Jintao administration downplayed the Three Represents, fearing that the inclusion of capitalists in the CPC would erode Party legitimacy; instead, official discourse increasingly referred to a 'people-centred' governing ethos as part of a broader shift towards ideology-based strategies of Party legitimation. 27 Xi Jinping's administration accelerated this process, with scholars suggesting that the Party's expanded role, coupled with the dismantling of Maoist-era bureaucracy indicated an 'advance of the Party in the wake of the state's retreat'. 28 Indeed, the Party's comprehensive role is explicitly stated in amendments to the Constitution, ratified at the 19th National Congress, and exemplified in the following passage from Constitution's revised preamble: The Party exercises overall leadership over all areas of endeavour in every part of the country. The Party must adapt to the demands of reform, opening up, and socialist modernization, remain committed to practicing scientific, democratic, and law-based governance, and strengthen and improve its leadership. The Party must, acting on the principle of guiding the overall situation and coordinating the work of all sides, assume the role of leadership core among all other organizations at the corresponding levels (emphasis added). 29 The implementation of this expanded role at the grassroots level meant creating Party branches within so-called 'new economic entities' (新经济组织, xin jingji zuzhi) and social organisations to ensure that both profit-driven and welfare-providing organisations performing roles previously conducted by government-led bodies would conform to political consensus. 30 Citizens' role in urban governance has also expanded from a solely passive or complaint-making form of political subjecthood to include one based in part on a type of active citizenship. 31 Political leaders consider participatory mechanisms important means to mitigate social conflicts caused by rising economic inequality and to promote both local stability and 'vitality'. 32 Thus, in the last 15 years, myriad new participatory initiatives have been implemented, creating extensive opportunities for citizens to engage in deliberative processes alongside government officials. 33 Citizens can contact local mayors through an online portal with questions and concerns. 34 In participatory pricing events, citizens, businesses and government officials come together to discuss and set the prices for key utilities and transportation services. 35 Citizens can use online mechanisms to evaluate the work of government departments 36 and rate the trustworthiness of local businesses, government departments and fellow citizens. 37 And, as we discuss below, PB initiatives allow citizens to provide input in the distribution of local budgets.
However, to ensure political stability, new participatory mechanisms must be integrated into the Party structure. Our article investigates how this is done in the case of PB. Consequently, we make three contributions to the existing literature on local governance in urban China. First, we challenge the bulk of the literature on PB, which tends to portray the practice either as an emancipatory innovation that empowers citizens to hold government to account or as a technocratic form of governance that makes public administration more efficient. Instead, by demonstrating PB's close relationship to both local government and local Party institutions, we conclude that citizen empowerment and bureaucratic optimisation are secondary goals to the task of ensuring ideological conformity, political stability and strengthening the legitimacy of the CPC. Indeed, the introduction of PB in China is, to follow O'Brien, 'less to do with responsiveness and changing state-society relations and more to do with state building, restructuring bureaucratic ties and making Party rule predictable and effective.' 38 Secondly, we demonstrate exactly how the roles of government and Party differ at the grassroots level, something that is rarely distinguished in Western scholarship. Yet we argue that this is essential to understand both the rationale for participatory projects in China, as well as their implementation. Finally, our article explores how local governing institutions are opened to citizen input through the adoption of globally promoted techniques of participatory governance and citizen engagement, but that these techniques take on very different meanings in China's complex political environment. What is a form of radical democracy in one context may strengthen sovereign power in another.
The article is structured in three parts. First, we summarise the state of research on PB in China, showing that the overwhelming majority of works focusses on the role of citizens in participatory initiatives. Second, we explore Chinese official discourse on participation and Party-building, showing that Party-building is presented as a tool to achieve 'citizens' orderly participation', and suggesting that this link requires us to focus on the Party's practical role in participatory initiatives. Third, we elaborate our two case studies, explaining state-Party-society dynamics in each. Although the two cases are very different, they illustrate how different political actorsa state organ and a social organisationfuse citizen participation with Party structures.

PB research in China: where's the party?
In its most basic form, PB is the idea that local citizens should have a say in how the city or district budget is spent. It was initially developed in the 1990s in the Brazilian town of Porto Allegre by the left-wing Brazilian Workers' Party, where it was utilised as a radical, emancipatory form of participation that could ensure public funds were under popular control. Following PB's success in Brazil, it was championed by the World Bank as one of 43 'best practices' of urban governance, and was described by the UN as a technique which could 'greatly enhance transparency and accountability and improve service delivery at the local level'. 39 Since then, PB has spread rapidly around the world: according to a 2018 count, there are over 3,000 PB projects in 40 countries. 40 The bulk of analyses of PB consider it as a paradigmatic case of global policy diffusion and assess the extent to which it is employed in specific instances either to serve citizen empowerment or to enhance technocratic governance. 41 As Biaocchi and Ganuza note, 'the seemingly endless adaptability of Participatory Budgeting to the most varied contexts and its compatibility with the most diverse political projects: left, right and center.' 42 The consensus appears to be that PB has evolved far from its Brazilian roots to serve as a mechanism that helps to implement a neoliberal programme of allocating scarce resources, minimising the role of central government and legitimating funding cuts.
In China, PB has become increasingly ubiquitous over the last fifteen years. Numerous initiatives have been instigated by a variety of local and international actors, including the World Bank and various foreign academics specialising in participatory governance, and hence take a wide variety of forms. Two PB initiatives in Wenling, Zhejiang Province, have formed the empirical focus of most studies of PB in China. 43 In Zeguo Township, citizens were randomly selected to participate in DP events over several days, while in Xinhe Township, citizens were invited to public hearings to engage in discussions about the local budget with the Local People's Congresses (hereafter, LPCs) in a manner similar to Minhang district described below. However, these are far from the only forms of PB in China. For instance, iterations of PB have also been uncovered in Chengdu, Sichuan Province, in which village representatives are elected to discuss the local budget at monthly meetings. 44 Space constraints prevent us from fully analysing all of China's PB experiments.
Analysts have characterised China's numerous PB experiments according to different 'logics' or types. He delineates three logics along which PB operates: an administrative logic, whereby PB is used to strengthen the bureaucracy; a political reform logic, whereby PB is used to enhance the effectiveness of LPCs and strengthen its deputies vis-a-vis the local government; and an empowerment local, whereby the delegation of power to citizens is the primary focus. 45 Similarly, Cabannes and Ming categorise three types of PB according to various organisational structures: first, a type directly inspired by the Porto Allegre model, whereby citizens participate in the decision-making, execution and oversight of part of a public budget; second, a type involving public discussion with LPCs; and third, a type whereby villagers elect a budgetary council and oversight group that collects proposals from villagers and monitors implementation of chosen projects. 46 A third tripartite model of PB practices that focusses on the role of citizens is presented by Yan and Xin: a model that utilises a new representative body to enhance citizen engagement; a model that allows citizens to participate in consultative deliberations that officials use in budgetary decisions; and a model in which information concerning the budget is made public to citizens in an effort to enhance transparency. 47 However, most analysts doubt the ability of Chinese PB to delegate power to citizens. Cabannes and Ming conclude that 'Most PB experiences in China, including the most innovative ones, are essentially urban-based and quite limited in scale. They are mostly consultative and are usually not fully open to the general public, or are limited to public hearings'. 48 Overall, analysts conclude that, although PB does partially unsettle existing governing arrangements, 49 PB in China is far from the emancipatory practice developed in Porto Allegre.
While our findings generally support these conclusions, we take a different focus since studies have not investigated the role of the CPC in PBa crucial factor that explains Chinese PB's difference in relation to its Brazilian iteration. While some research points to the strengthening of LPCs through PB, most studies focus on citizens' relationship to the 'party-state', which blurs these two parts of the governing apparatus. Yet taking the Party into account is important because, as we show below, the presence of the Party determines the character of civic participation offered through PB. One of the only studies to explore the effects of participatory governance specifically on government institutions in China is by Pavli cevi c, who explored participatory mechanisms in Zhejiang Province, including public hearings, citizen consultations and public surveys, arguing that 'public participation in these non-electoral mechanisms exerts impact on party-state's policy-making and performance while also simultaneously strengthening its governing capacity'. 50 He suggests that, contrary to common assumptions in Western literature, projects aimed at strengthening the state and those seeking to enhance citizen participation do not contradict one another. We concur with Pavli cevi c, but expand his analysis to explore how PB initiatives in particular enhance relations between citizens, government and Party at the sub-municipal level.

Participation and party-building in official discourse
This section explores the way in which citizen participation and the Party are brought together in government discourse. We suggest that the central terms for understanding the connection between the two are 'citizens' orderly participation' (公民有序政治 参与, gongmin youxue zhengzhi canyu) and 'Party-building' (党建, dangjian), with the latter invoked as a strategy to ensure that the former remains 'orderly'. While all citizens are exhorted to engage in participatory governance, processes of Party-building aim to install a well-trained, professionalised core of cadres in every institution, as well as in participatory activities themselves, both to present the Party as engaged with societal issues and to ensure political conformity and, consequently, social stability.
China's leaders have promoted the concept of 'citizens' orderly participation' to conceptualise a third avenue of state-society relations that would occupy space between citizens' disengagement from governance, on the one hand, and more autonomous and potentially disruptive forms of participation, such as petitioning (上 访, shangfang) and protest, on the other. According to He, PB is a key example of discourses of 'citizens' orderly participation' in practice. 51 However, much like the concept 'Party-building', the precise meaning of the term is hard to pin down. It was first mentioned in October 2000 during the Fifth Plenary Session of the CPC 15th Central Committee, which called for 'enhancing the democratic political system, promoting scientific and democratic decision-making, and expanding orderly public participation', and defined the role of government to 'lead citizens to manage themselves according to the law'. 52 This was the first time in which citizens' self-governance, as part of broader processes of bureaucratic transformation, was explicitly linked to citizen participation. Two years later, at the 16 th Party Congress, Jiang Zemin called for the enhancement of the 'vitality' of the Party, which would 'develop diverse forms of democracy, expand citizens' orderly political participation, and ensure that the people go in for democratic elections and decision-making, exercise democratic management and supervision according to law'. 53 At the 17 th Party Congress five years later, Hu Jintao stated that 'citizens' participation in political affairs will expand in an orderly way'. 54 In 2014, shortly after coming to power, Xi Jinping called for grassroots participatory mechanisms to be strengthened, emphasising the need to 'expand people's democracy by improving democratic systems, enriching forms of democracy, and creating more channels for the practice of democracy, and enable broader, orderly political participation of citizens at all levels and in all domains,'. 55 However, by the 19th Party Congress in 2017, Xi had subtly but significantly altered his conception of the political nature of participatory governance: 'It is necessary to improve the Party's leadership and governance methods to ensure that the party leads the people to effectively govern the country. We will expand the people's orderly political participation to see that in accordance with law they engage in democratic elections, consultations, decision-making, management, and oversight'. 56 The shift in Xi's rhetoric from the use of 'citizens' (gongmin 公民) to 'people's' (renmin 人民) demonstrates the broader shift away from citizen participation as a key locus of political legitimacy. 57 Instead, by placing Party leadership before 'people's' participation, the Party has become the primary locus of political agency, and the possibility for citizens to provide autonomous political input through participatory mechanisms has been deemphasised. Indeed, the trends we observe in Minhang District and Yangjing Subdistrict appear to confirm the implementation of this discursive shift: the 'orderly' nature of participation through PB is maintained through the integration of participatory mechanisms into the Party's grassroots network.
In order to understand how the process of integration is envisaged, we consider the ubiquitous in Chinese official literature yet somewhat oblique term, 'Party-building'. Nie and Wu define Party-building as 'a top-down process of political assimilation and institutional isomorphism'. 58 While we do not disagree with this, given that Party membership stretches across the state-society divide, it is useful to distinguish Partybuilding that occurs in political institutions, in the 'state', and that which occurs among citizens and NGOs, in 'society'. Indeed, in official CPC discourse, Party-building generally refers to a series of activities undertaken by the Party in order to strengthen its overall leadership, both at the macro level, in government institutions, and at the micro level, in grassroots organisations. 59 Regarding Party-building in government, Xi emphasised that government bodies should 'allocate a central role to the Party Committee, build the team and serve the people, thus promoting a deeper integration between Party-building and government activities'. 60 Although Party and government are formally separate institutions, in practice their relationship overlaps. For instance, leadership positions within government institutions are occupied by Party members; Party committees exist within government departments at all levels, in order to monitor and guide their activities; and government officials are required to uphold CPC ideology, and training sessions are provided in government departments. Xi has further merged the functions of government and Party by incorporating several government departments into Party apparatus, including the State Office of Public Reform and the State Bureau of Public Servants.
Xi has placed special emphasis on Party-building at the grassroots level, stating that 'grassroots organizations are the "nerve endings" of the Party's body … Only when it reaches the ground can it take roots, and the deep roots can grow leaves'. 61 Activities pertaining to Party-building in local social organisations were set out in the 'Opinions on Strengthening Party Building Work in Social Organizations', issued by the General Office of the CPC Central Committee in September 2015. These include: publicising and implementing the resolutions of higher level Party organisations; encouraging Party members to study Xi Jinping Thought and other ideological documents; organising cultural activities that foster a positive atmosphere; providing opportunities for talented individuals to advance their careers in alignment with the Party; ensuring that all Party members fulfil the obligations of their Party membership' and maintaining relations with the Party organisations of other grassroots organisations, such as the Communist Youth League, trade unions and the Women's Federation. 62 Despite this focus on strengthening Party leaders, growth in CPC membership has slowed under Xi and research indicates that it has become increasingly difficult to join the Party, which suggests a preference for highly qualified, professional individuals. 63 Thus, Party-building either at the grassroots or the state-level does not necessarily refer to the expansion or growth of the CPC in terms of membership numbers, but to the increasing influence of the Party over different aspects of social and political life, including guiding and monitoring non-Party members, both in institutions and in wider grassroots society. 64 Government-organised participatory activities assist with this because they bring Party members into closer contact with citizens, enabling them to promote Party ideology and to ensure that participatory activities correspond to the Party line.
This discourse of citizens' orderly participation has filtered to the municipal level, with both government and Party officials calling for an expansion of new mechanisms of participatory governance. The 2012 report from the Shanghai Municipal government pinpointed grassroots participation as a solution to problems of urban governance. In 2014, social governance innovation became Shanghai's 'number one topic', which trained an initial batch of 43,000 grassroots community personnel, decentralising decision-making powers to the street level on a range of topics. 65 In 2017, Shanghai's outgoing Party Secretary, Han Zheng stated to the 11th Municipal Party Congress that a key task was to 'consolidate and develop socialist democracy and expand the orderly political participation of citizens'. 66 In 2020, Resolution of the Ninth Plenary Session of the 11th CPC Committee called for the creation of 'a city where everyone can participate in governance in an orderly manner'. 67 Thus, the expansion of grassroots participatory mechanisms in Shanghai can be seen as an operationalisation of this national-level discourse. We now elaborate two such mechanisms.

Participatory budgeting in Minhang and Pudong
This section explores two projects that involve citizens in budgetary decision-making in Shanghai: the Minhang District People's Congress and a Residents' Committee in Yangjing Sub-district, Pudong. These cases provide an example of a district state-facing PB initiative and a grassroots society-facing PB initiative. Shanghai was selected as our research site because it is well-known as a leader in innovative governance, and experiments that begin in this city are often used as examples of best practice elsewhere (indeed, this is what happened in the case of Minhang PB); hence, understanding how participatory governance is managed in Shanghai may help us understand how other Chinese cities could approach this problem in the future. In each case, we collected reports from relevant administrative websites and WeChat channels regarding each PB initiative. These materials enabled us to build a picture both of the procedure itself, and of the way it has been presented to the public as an example of deliberative democracy. We supplemented these online resources with 10 in-depth, semi-structured research interviews conducted between 2019 and 2021 with local officials, Party-members, citizens and members of community foundations, using them to corroborate and supplement the information gathered online, as well as to understand their views of the process. Our research shows that, although the two practices of PB differ substantially, the CPC is intimately involved in both, engaged in direct supervision in the first and more subtle penetration in the second. Our two cases demonstrate how PB is managed by the Party and contributes towards the wider Party-building project.
Minhang district: PB in the district people's congress Local People's Congresses (LPCs) are public monitory bodies that cascade in pyramid form from the National People's Congress down to provincial, municipal and district congresses. LPCs' responsibilities involve supervising local government and judicial bodies, drafting legislation, ensuring that the Constitution is adhered to, conducting oversight of local Party activities, and overseeing the distribution and implementation of the local budget. 68 Although the government and the LPCs are formally distinct, with the latter envisaged as a sort of popular assembly, there has been substantial overlap of their functions in practice and both are seen as 'state organs' according to the Constitution. LPCs are also sites where Chinese citizens can engage with electoral politics: deputies of district and sub-district levels of congress are elected by the public in a secret ballot. At the same time, the CPC is strongly involved in LPCs, through the existence of a party group within each LPC that provides ideological guidance (see Figure 1) 69 and by ensuring that candidates for election are aligned with the Party. Delegates to LPCs must be recommended for election by the local CPC branch, which led to a high number of CPC members in the congresses, although in the past some LPCs have sought to 'limit' the number of CPC members to 65%. 70 The blurring of functions of government and LPC was seen to breed corruption and inefficiency. Hence, in 2007, a law was passed clarifying the roles of the two institutions, which as Alm en argues, served to reinforce Party control of LPCs, since it sought to improve cadre quality which in turn would strengthen the governing capacity of the Party. In other words, LPC empowerment was an intentional way of strengthening the Party and simultaneously limiting Local Party Committees' arbitrary power. 71 Thus, the Party's influence over Congress seeks to bring state and society closer together by cementing the connection between the 'will of the people' and the 'will of the Party'. He has argued that the introduction of PB into LPCs has been interpreted as a way of enhancing the effectiveness of LPCs, and to empower LPC deputies vis-a-vis the government. 72 But what role does the Party play in its PB initiative? This section probes this question in the case of Minhang District People's Congress.
Minhang District is located on the outskirts of Shanghai. It is mostly a residential district, but is also home to factories and production facilities. Covering 372.56 km 2 , it administers 4 subdistricts, 9 towns and one industrial zone. In total, it has 132 village committees and 427 RCs. Enhancing participatory governance and strengthening the Party have been important priorities for the district's management, and has been articulated by district government, Party and People's Congress. For instance, Minhang District's 2018 Report on the Construction of Rule of Law Government emphasised the need to 'strengthen the construction of grassroots systems, consolidate the legal basis for grassroots autonomy, and explore innovative co-construction and shared social governance models'. 73 In 2020, the Secretary of the District Party Committee emphasized the link between citizen participation and political alignment, stating, 'We must actively explore effective ways to make full use of the democratic channels of the People's Congress, continue to expand the orderly political participation of the public … At the same time, the work of the National People's Congress should be incorporated into the overall work layout of the district … and to give full play to the political leadership of the Standing Committee of the National People's Congress'. 74 In 2021, Minhang District People's Congress published a report stating its aim to 'form a high-level governance system compatible with the main urban areas of modern international metropolises, realizing the vision that "everyone has the opportunity to make a difference in life, everyone can participate in governance in an orderly manner, everyone can enjoy a quality life, everyone can truly feel warmth, and everyone can have a sense of belonging"'. 75 Minhang's PB initiative is unusual because it occurs at the district level: it involves all the district's subdistricts and towns, as well as the industrial zone (most PB initiatives in China occur at the sub-district or street-level). District Party Secretary, Sun Chao, initiated a pilot project in 2007 following the legislative changes to LPCs, in cooperation with well-known legal scholar, the late Cai Dinjian.
Following several years of refinement, in 2012, delegates at the Fifth District People's Congress passed a resolutionthe first of its kind in Chinastating that the government budget expenditure should be overseen by the District People's Congress. 76 Thus, from the start in Minhang, PB was a project instigated, guided and overseen by the leading Party official in the district. Figure 1 illustrates the relationship between the Party, state and citizens in Minhang's PB initiative.
The process of PB in Minhang consists of two key elements: the performance evaluation carried out by the District Government's Finance Bureau and regular public hearings held by the District People's Congress (DPC) Standing Committee. Every year, the Finance Bureau evaluates potential programmes to be funded through the district budget, and selects those that will be discussed in a public hearing. This process includes the participation of DPC deputies and professional experts invited by the government. The District Finance Bureau conducts the pre-evaluation of the performance evaluation of these projects, considering their feasibility and necessity, and gives a numerical score to each project. Projects that are more expensive, with a wider reach and greater levels of potential controversy are more likely to be selected.
Public hearings are announced online and are where the citizen participation occurs, forming the central part of Minhang's PB process. 77 Participants are composed of speakers (chenshuren) and listeners (pangtingren); the former comprise relevant government department representatives, deputies from the DPC, ordinary citizens and other relevant government officials. Public speakers are either invited or apply by themselves; those who apply should state their views clearly beforehand by writing down their opinion in the registration form downloadable from the website (including basic information such as name, address, ID number and job, etc.). The position of speaker is open to all citizens living or working in Minhang district aged 18 and over. Following confirmation of the speakers, the DPC Standing Committee holds a meeting before the formal hearing, in which speakers present their planned interventions. Officially, this is in order to avoid duplication or deviation from the subject of hearing, although a second reason for the meeting, one can infer, is to ensure that opinions conform to the political line: for instance, speakers are encouraged to give constructive suggestions on the execution of the budget, rather than voicing total opposition.
The hearing proceeds in three parts. First, the speakers give presentations in turn: the proposal is introduced by its initiator in government, followed by interventions from the public speakers and expert speakers. Secondly, a twenty-minute questionand-answer session takes place, where the public speakers put questions to the speaker representing the government. The final section involves a debate, and is based on the earlier presentations. It usually involves two rounds: first, the government speaker responds to the viewpoints presented by public speaker; and second, public speakers debate the views of the government speakers. The DPC Standing Committee stipulates that presentations made by government departments should be clear and comprehensible: complicated financial statements should be converted into ordinary language, intelligible to DPC deputies and listeners from the public. 78 A full list of projects debated by public hearing are listed on the Minhang DPC website, but recent examples include the maintenance of sewage pipelines, the provision of a bicycle renting service, and subsidies for nursing homes. 79 Sums allocated to particular projects number in the millions of yuan; the most expensive recent project (2010)(2011) was costed at 84 million yuan and funded agricultural subsidies to local farmers. 80 Further close connection between the Party and Minhang's PB initiative is exemplified in the biographies of participants. For instance, a citizen participant, who was also the secretary of the local branch of the Communist Party Youth League, told us how he had participated in 'the first batch of party-building training courses for social organisations in Shanghai' before attending at least 10 rounds of Minhang's PB hearings. 81 Similarly, in 2015, Shanghai Municipal People's Congress showcased the work of a long-running member of Minhang DPC, Qian Tianxin. He stated that as a member of the CPC and a DPC delegate, he should 'serve the people heart and soul'also the fundamental stated purpose of the Party. Qian was frequently invited to give lessons to Minhang District Party branches and was invited to participate in numerous public hearing on the budget as a public speaker. From the perspective of citizens, he was considered a reliable person who can articulate demands on their behalf; from the perspective of the government, he is able to represent local issues effectively. 82 These two personages embody the ideal figure for PB in Minhang, who is able to 'shuttle' smoothly between Party, state and society.

Yanjing Sub-district, Pudong: PB in the Residents' Committee
The PB experiment in Yangjing Sub-district was led by a social organisation, Yangjing Community Foundation, in collaboration with a local RC, Yuyang RC. The Party is active in both organisations, with the community foundation exhibiting the organizational attributes of the Party and the head of the RC also being a Party member. Together, this enhances the influence and mobilizational ability of the Party at the grassroots level. This subsection introduces the two organisations, explaining how the Party operates through them, before describing the PB initiative.
In many respects, Yangjing Community Foundation is similar to private social organizations in China: it has an independent legal status, a complete board of directors, board of supervisors and application teams, and it is managed by the Civil Affairs Bureau. But the Community Foundation's distinguishing feature is that it is fully financed by the Yangjing Subdistrict Office. 83 This sets it apart from traditional social organizations, which are funded by private associations or interest groups. In addition, the deputy investigator of the Yangjing Sub-district Office also served as one of the Community Foundation's directors. In the community activities enacted by the Foundation, the Sub-district Office has also given it full support: for instance, the Subdistrict Office helps the Foundation to cooperate with RCs under the Office's jurisdiction. The director of the Yuyang RC described the relationship between the foundation and the Sub-district Office: The Foundation and the Sub-district Office are closely related and the Foundation cannot be regarded to be completely independent. After all, a lot of work is still subject to the Sub-district Office's leadership and the Foundation cooperates with the Subdistrict Office. 84 Furthermore, there are deep connections between the Foundation and the Party. The Sub-district Party Working Committee provides the Foundation with office and activity space, located in the Yangjing Sub-district Party and Mass Service Centre (dangqun fuwu zhongxin). It shares an office with the Party's mass organizations, including the Communist Youth League's Working Committee. Importantly, the Foundation has a unique relationship with Yangjing's Party Working Committee. First, prior to the Foundation's establishment, its now secretary general had served as deputy secretary of a local Communist Youth League Working Committee, and his political affiliation as a CPC member is clearly stated on the web page. 85 In addition, the Foundation's annual report indicates that there is a Party organization within the Foundation. 86 However, since the number of Party members in the Foundation is fewer than three, a Party branch for the Foundation cannot be established independently. Hence, it has formed a Party branch with other social organizations in the subdistrict called 'The Second United Party Branch of Yangjing Social Organizations'. This branch is subordinate to the Yangjing Sub-district Party Working Committee.
The Foundation's complex financial and personnel links with government and Party make it difficult to characterise as an NGO in the Western sense; however, it cannot be regarded as a Party or government organisation either since it works alongside other foundations and community groups in projects commonly enacted by welfareproviding NGOs. Thus, although its bureaucratic structure links it closely to the state, its activities are similar to other more independent organisations. 87 Hence, we consider the Foundation to have 'two faces'one looking in towards official structures and one looking outwards towards citizens and social organisations. Indeed, this is a relatively common feature of grassroots organisations in China.
RCs are the lowest administrative unit in urban China, managing social and political life across four to five housing estates. Although they have existed since 1950s, their role was transformed in 1980s, when the socialist work units, (danwei 单 位 ), which delivered welfare to employees, were dismantled and most of their functions shifted to RCs. Thus, these committees were tasked not only with service provision, but also with Party-and community-building activities. This role is reflected in the Ministry of Civil Affairs document, published in 2000, outlining community-building as a 'crucial tool in national efforts to promote social development, to raise living standards, to expand grass-roots democracy and to maintain urban stability'. 88 RCs should do so by involving local residents in participatory activities. Like the Community Foundation, although RCs are formally autonomous grassroots organisations, in practice they are closely linked to Party and government, since the head of the RC should be a Party member, and the local government Subdistrict Office manages and coordinates their activities. 89 Yangjing Sub-District is located in northwest Pudong New District, east of the Huangpu River. It is one of Shanghai's most modern districts, headquartering many global corporations. At the end of 2019, the registered population was about 115,200. There are 40 RCs, managing 116 residential communities. 90 Over the past 20 years, with the development of Pudong New Area, Yangjing Sub-District has developed into a plural community including traditional households of Pudong, relocated households from across the river (Puxi district), newly constructed households, and the households of migrant workers and international residents. Since Yangjing Sub-district Office felt unable to meet the needs of this increasingly diverse demographic, in 2013 it founded the Yangjing Community Foundation as a pilot project to drive forward innovation in grassroots governance. The Community Foundation and the Yuyang RC were the main organisers of the PB project in Yangjing.
Local experiments in PB began in 2004 with a one-off experiment in Huinan Township, in which a scoring method was used to gauge residents' support for various projects. 91 Later, in 2011, Lujiazui sub-district pioneered the competitive allocation of the district government's 'Self-Governance Fund', a pot of government money whose use can be decided independently by local government Sub-District Offices and local RCs. To compete for the money, the Sub-District's RCs were required to conduct broad consultations with citizens, and then propose fully costed projects which are first assessed by a government panel and then presented to randomly selected citizens for public deliberation. This change aimed to encourage RCs to engage community residents in local governance, since one of the biggest problems for grassroots governance is citizens' lack of interest in community affairs. 92 The result was successful in engaging the residents and Pudong New Area Civil Affairs Bureau went on to promote the Self-Governance Fund in other sub-districts and RCs. However, the district Civil Affairs Bureau felt that the initial iteration of the Self-Governance Fund had limitations: in particular, the final use of monies allocated by fund was still determined by RC staff, involving just a small number of core community members. 93 Therefore, in 2017, Yangjing Sub-district Community Foundation implemented a PB project that engaged all residents in voting in cooperation with Yuyang RC, which manages 1553 households. 94 The total amount available was 30,000 RMB, half of which came from the district government's Self-Governance Fund, and the other half was raised by the Community Foundation by soliciting residents' donations. Figure 2 illustrates the relationship of the PB project to the Subdistrict Community Foundation and Yuyang RC. The PB process proceeded as follows. Firstly, the Community Foundation invited interested residents and staff from Yuyang RC to training sessions led by two PB experts from Taiwan. Then, with support from the Community Foundation and the RC, residents formulated their budgetary plans in detail and made leaflets and banners to advertise their projects around the local estates. In addition, the Community Foundation and RC mobilised residents to participate in the voting by organizing a group of middle-school student volunteers to assist residents with advertising and voting, setting up information tables in the communal park and going door-to-door to collect residents' votes on which project should be funded. According to the Sub-District Foundation, the school students 'think more actively and made some propaganda materials for the project, even in the form of cartoons for children.' 95 In total, 965 of the 1553 households voted, with turnout reaching 62%, far exceeding the expected number (the RC and the Foundation had expected 20%-30% turnout). According to the Director of the Community Foundation, the middle school students were more successful in eliciting public participation than the traditional official residential leaders. 96 Two main projects were up for deliberation, each advanced by two different community groups. One project was proposed by a group of parents, active in Yuyang community for years, advocating protection for abandoned pets and educating citizens on responsible pet ownership. The second project was proposed by a group of elderly residents skilled at Chinese paintings, who had organized an on-going Calligraphy and Painting Club. This project proposed to re-paint the corridors of several buildings and decorate them with mounted paintings by club members. In the competition between the two projects, the 'Corridor Redecoration' project won 78% of the votes, and was implemented.
Finally, due to the Foundation's close relationship with the Sub-District Office, both the RC and the residents have more trust in the activities it organises. This is exemplified in our interview with Yuyang RC: For residents, these [activities] are all seen as being organized by the Sub-District Office. When residents come to participate in activities, they think they're coming to the Sub-District Office. In fact, residents are more at ease with participating in such activities because they are organized by the Sub-District Office. If it is organized by a social organization, it is not so easy for residents to participate … There are still some political sensitivities behind it. 97 This highlights how the close relationship to government provides legitimacy for the Community Foundation. In addition, the Foundation's embedding into the local governing ecology avoids a political problem that may appear in grassroots governance, where social organizations become too powerful and replace the functions of grassroots Party organizations. 98 Overall, the Foundation's PB experiment demonstrated that social organizations can form a strong cooperative relationship with grassroots administrative organizations, such as RCs, strengthening their ability to mobilise citizens, implementing the political agenda of enhancing participatory governance, and ensuring political allegiance at the grassroots level. However, despite the initiative's success in increasing citizen participation, the PB project was discontinued. The Foundation's Secretary-general explained in our interview that their PB project was only an experiment: it required too many resources, and after 2017 the Self-Governance Fund returned to distribution by the RC. 99

Discussion
Our cases have explored the crucial process by which the Party permeates two types of PB initiative: one state-facing and one society-facing. Our study reveals four key, inter-related differences in the role of citizens, Party and government. These pertain to the overall aim of the PB initiative, the nature of the budget that is opened up to citizen deliberation, the way in which citizens were engaged in the process, and the Party's role in the execution of the initiative.
Firstly, in Minhang, PB was instigated to enhance the authority of the DPC vis-a-vis the district government; in Yangjing, PB aimed to resolve the problem of low citizen engagement in local governance. In Minhang, PB was introduced as a local innovation alongside a raft of national reforms, aiming to clarify the responsibilities of LPCs and local governments; PB contributed to this, enabling LPCs to conduct budgetary oversight and enhance citizen participation, thereby increasing the legitimacy of the LPCs' scrutiny. In Yangjing, while the PB experiment also responded to the government's broader aim of enhancing civic participation, it specifically sought to improve the distribution of Pudong New Area's 'Autonomous Fund', a targeted reform aiming to improve the efficiency of the scattered government-allocated subsidies for community activities. Yangjing's PB experiment thus aimed to solve the dual problems of the lack of public opinion in the distribution of the Self-Governance Fund and the low level of residents' participation in community activities. Thus, compared to Minhang, Yangjing's PB experiment focused on a specific, localised problem and consequently adopted more flexible activities, compared to the formal PB hearings enacted by Minhang District's higher administrative level.
This more formal approach to Minhang's PB is underscored by the huge contrast in the nature of the budget under deliberation: Minhang's budget comprises the District budget allocated by the government, while in Yangjing, the budget is partly allocated by the government through the Self-Governance Fund and partly raised by citizens' donations. Thus, while Minhang's public hearing involves the allocation and oversight of substantial sums of public money totalling several million RMB, in Yanjing, the amount is marginal and is partly raised by donations from citizens themselves. In the latter case, this means that citizens feel that they have a greater stake in the distribution of funds, and are hence more likely to participate as they want a say in how their money will be used.
This brings us to our third difference: in Minhang, citizens must apply to participate in the PB hearings and their applications and views are vetted by the DPC; in Yangjing, all citizens were effectively mobilised into participating. In Minhang, although the hearing is open to all district residents, few people actually participate; those who do normally have a direct stake in the outcome of a particular discussion. According to an interviewee, just five or six ordinary citizens apply to attend each public hearing, with two or three turning up on the day. 100 In Yangjing, the Sub-district Foundation deployed its wide social network to encourage all residents to participate in the project, from RCs to residential leaders (louzhang 楼长), even enlisting middle school students. This contrasts with the promotion of Minhang's PB programme, which only appears on the traditional government websites and government-affiliated media. In addition, the highly controlled nature of participant selection in the case of Minhang demonstrates the way in which PB organisers manage the risks of including citizen voices and ensure that the desired outcome of budgetary deliberations is achieved. In the case of Yangjing, the political risk is much lower since the PB pertains only to monies intended for the housing estates, and the overall goal of PB is to engage citizens rather than apportion scarce resources.
Finally, and most importantly for this article, in Minhang, Party involvement was formal, with the Party District Secretary directly involved in instigating PB and Party groups involved in overseeing it; while in Yangjing, Party was more subtly embedded in the Foundation and the RC. However, even though in Yangjing's case PB was not a direct Party project as it was in Minhang, the close informal relationship of the Sub-District Foundation to Party organisations is exemplified in the biographies of Foundation staff. This comparison shows how the CPC works differently in different types of state-society interactions, ensuring that citizens' participation and engagement with local authorities remains 'orderly'. In the state-facing PB, Party oversight was easy to establish because formal Party organizations function alongside district government offices and Peoples' Congresses, while at the grassroots level the line between government, Party and society becomes much more blurred and individuals assume several roles at the same time, especially when it comes to social organisations. This fact of Party penetration into PB is not to dismiss the sense of meaning that the process might instil within citizens: research has shown how local and grassroots participatory activities in Shanghai contribute towards feelings of empowerment in citizens. 101

Conclusion
This article has shown how a shift in the role of the Party vis-a-vis government and society, exemplified in Xi's speeches since 2017, is operationalised at the sub-municipal level in processes originally conceived to place budgetary decisions in the hands of citizens. This central role of the CPC in PB contrasts sharply with the way in which PB is enacted elsewhere. In China, Party organisations penetrate into both state organs and societal organisations, and the participatory activities they initiate, coordinating their activities and strengthening itself in turn. This process of 'Party-building' can happen formally in cooperation with state organs, as in the case of Minhang District, or informally, through social organisations and grassroots networks, as in the case of Yangjing Sub-District.
We conclude this article by returning to the three contributions outlined in our introduction. First, we have shown that in our two cases, PB's primary intended function is not to 'emancipate' citizens: at the level of practice, the Party maintains extensive control over the PB process, ensuring that civic participation remains 'orderly', and at the level of discourse, language evoking notions of empowerment and emancipation are not used in the materials posted online about the PB events. Neither is PB used to legitimise 'neoliberal' reforms: PB adds to, rather than streamlines, existing architecture of local bureaucracy, and, rather than devolving responsibility for financial decisions onto citizens, the Party supervises citizens' participation from above. Instead, we have suggested that the primary overall aim of PB is to elicit citizens' orderly participation, that is, participation aligned with the Party agenda, thereby enhancing Party leadership in its corresponding administrative level. This suggests that prevailing conclusions about the function of PB need to be re-thought when it comes to its implementation in urban China.
Secondly, we have distinguished the roles of the Party in specific sub-municipal PB projects, one in government and one in society, describing how they connect and illustrating these connections in two detailed diagrams. In Minhang, District government officials are involved in public hearings, alongside DPC delegates, invited experts and the occasional citizen; however, it is the DPC, and specifically Party members within the DPC Party group who guide the process and ensure it conforms to Party norms. In Yangjing Sub-district, the government institutionthe Sub-District Officeprovides the institutional contacts and resources to the social organisation implementing the PB project, while the social organisation's informal connections to the Party ensures it holds legitimacy in the eyes of the RC and the residents themselves. In both cases, the role of government is secondary to that of the Party when it comes to the execution of the PB initiative. Thus, our research contributes to the small pool of existing research in English that treats Party, state and society as analytically distinct. This leads to several important questions for further research: how does PB impact internal Party relations? And to what extent does PB influence the view of the Party in the eyes of citizens? Although we could not address these questions above, we hope we have opened up these avenues for further study.
Finally, we have shown that practices of participatory governance in China's oneparty state take on very different meanings to their iteration in liberal democracies, which deploy citizen participation to increase popular control of government processes. In China, we have shown that the opposite is the case: the Party manages participatory governance from above to ensure that outcomes of citizen deliberations fit the Party agenda. Notes 1. Two positions exist on the correct translation of zhongguo gongchandang (中国共产党).
In English, CCP is most commonly used; however, CPC is the translation officially approved by the Party itself. Debates over appropriate usage of these two translations are on-going amongst Chinese scholars. In this article we adopt the official translation. 2. We use 'social organization' as the literal translation of shehuizuzhi (社会组织). In the CPC Central Committee's 'Opinions on Strengthening the Party Building Work of Social Organizations', social organisations 'mainly include social groups, private non-enterprise units, foundations, social intermediary agents, and community organizations in urban and rural communities' (People.cn 2015). This is mirrored in some Western literature: Saich (2000: 124) refers to social organisations as 'both the more autonomous organizations and those set up by state agencies specifically to carry out social welfare functions'. 3. Fishkin, Democracy and Deliberation.

Disclosure statement
No potential conflict of interest was reported by the author(s).

Funding
This research was funded by a British Academy Postdoctoral Fellowship and an Exeter-Fudan Fellowship.