China’s Overseas NGO Law and the Future of International Civil Society

ABSTRACT China’s law to control international non-governmental organisations (INGOs) has sent shockwaves through international non-governmental organisations (NGOs), civil society and expert communities as the epitome of a worldwide trend of closing civic spaces. Since the Overseas NGO Management Law was enacted in January 2017, its implementation has seen mixed effects and diverging patterns of adaptation among Chinese party-state actors at the central and local levels and among domestic NGOs and INGOs. To capture the formal and informal dynamics underlying their mutual interactions in the longer term, this article employs a theory of institutional change inspired by Elinor Ostrom’s distinction between rules-in-form versus rules-in-use and identifies four scenarios for international civil society in China – “no change,” “restraining,” “recalibrating” and “reorienting.” Based on interviews, participant observation and Chinese policy documents and secondary literature, the respective driving forces, plausibility, likelihood and longer-term implications of each scenario are assessed. It is found that INGOs’ activities are increasingly affected by the international ambitions of the Chinese party-state, which enmeshes both domestic NGOs and INGOs as agents in its diplomatic efforts to redefine civil society participation on a global scale.

In late December 2014, following months of investigations into the operations of INGOs in China as part of a national security initiative, state media announced that the National People's Congress was deliberating a new law to regulate the activities of overseas NGOs (ONGOs) in China (Reuters, December 22, 2014). 1 The report also suggested that the law was not being drafted by the Ministry of Civil Affairs, which had been vested with management authority for both domestic NGOs and INGOs to date, but rather under the auspices of the Ministry of Public Security (MPS). Such reports sent shock waves through non-profit communities in China and abroad. INGOs, many of which had long operated without formal registration in legal grey areas and in close relationships with local Chinese partners, feared being banned as illegal entities. Foreign diplomats and third-sector representatives started lobbying the Chinese government to prevent at least the most draconian stipulations of the new law. Domestically, local governments and NGOs reacted with trepidation as they expected Public Security officials to scrutinise the symbiotic collaborations they had entered into over the years to improve social services, education, healthcare, environmental quality and other living conditions for local communities. The fact that many of them had worked with and received financial support from INGOs meant they could be seen as complicit in the shaky legal arrangements. To avoid being implicated in the expected public security scrutiny, some local actors limited or even ended collaborations in pre-emptive obedience. In retrospect, international and domestic actors changed their behaviour in anticipation long before the Law on Administration of Activities of Overseas NGOs in Mainland China (Zhonghua renmin gongheguo jingwai feizhengfu zuzhi jingnei huodong guanli fa, hereafter ONGO Law) eventually came into effect in January 2017.

From Trepidation to Adaptation: Coping with the Overseas NGO Law
Since the promulgation of the ONGO Law, many stakeholders have found ways to cope with it. While onerous stipulations have greatly intensified the bureaucratic burden on project managers, the MPS's implementation practice has been more accommodating than initially expected. Both the quantitative registration data made available since 2017 and qualitative insights from the ground indicate that the ONGO law has not led to an immediate exodus of INGOs from China (MPS 2020). Instead, the ambiguities and internal contradictions in the new regulatory environment for civic organisations in the Xi Jinping era (of which the ONGO law is only one part), indicate more creeping, longer-term and informal changes in the attitudes of INGOs, domestic non-profits and central and local party-state authorities. In our view, it is only by analysing these broader and longer-term dynamics that we can accurately gauge the ONGO law's true implications and the prospects for international civic actors in China.
To account for the substantial divergences between one-off changes in the formal-legal framework and more gradual, ongoing shifts in informal regulatory approaches and the perceptions and behaviours of the actors involved, we employ an approach informed by theories of institutional change, in particular by the distinction between rules-in-form and rules-in-use introduced by Ostrom (2005Ostrom ( , 2011. Based on her and on Cole's (2017) work, we deduce a set of four scenarios which imply different configurations of formal and informal rules. In our empirical analysis, we look for evidence to support or discard these scenarios and discuss the respective mid-term and longer-term implications for INGOs.
This approach contributes to the current academic debates on civil society in Xi's China and on the "shrinking spaces" globally for NGOs that operate internationally. On the one hand, we place the ONGO Law in a broader institutional context and emphasise the interactions between formal and informal changes in the relationship between the party-state and society at large. On the other, we scrutinise how new domestic regulations have led to gradual and subtle changes in INGO behaviour on a global scale. Straddling the domestic (central and local) and the global levels, the study aims to understand how the international ambitions of the Chinese party-state under Xi have affected its interactions with INGOs and their domestic counterparts and to develop a set of scenarios to outline the longer-term implications for civil society activism from a transnational perspective. In disciplinary terms, our scenario analysis aims to enrich the prevailing comparative politics perspectives in the analysis of "shrinking spaces" for INGOs by exploring how formal and informal changes in the domestic realm can interact with international-level agendas. At the same time, we deploy our theoretical and empirical insights on institutional change in China to engage with the international relations debate on global civil society and struggles over norms of civic participation in the twenty-first century (Toepler et al. 2020;Lang 2019).
The rest of this article proceeds as follows: after a review of scholarly interpretations of the ONGO Law, the theoretical section introduces the institutional change approach, in particular the conceptual distinction between rules-in-form and rules-in-use made by Ostrom and other institutional economists. From this theoretical perspective, four scenarios for the future development of international civil society and its interactions with domestic civic actors in China are identified. In the ensuing empirical analysis, evidence to identify each scenario's underlying driving force and assess its plausibility is presented. The concluding section weighs the likelihood and implications of the four scenarios in the longer term, while also considering external scope conditions. The empirical analysis relies on a broad range of evidence, including the rich secondary literature on state-society relations in China with its particular focus on central-local relations and the variegated triangular relationships between local governments, domestic associations and INGOs; the international relations debate on "closing," "shrinking" or "changing" spaces for civil society; the primary analysis of Chinese legal texts and policy documents; and our own field research. During the period between 2016 and summer 2020, over multiple field research stays, the authors not only interviewed domestic and foreign experts and practitioners but also used symposia and conferences in the non-profit realm as opportunities for participant observation. The formal interviews, conducted both in person and remotely, as well as the events used for participant observation and informal conversations are listed in the Appendix and referenced as Int01 through to Int28 for interviews and fieldwork sites as E01, E02, and so on for events throughout the text to safeguard participants' anonymity.

The Overseas NGO Law in the Academic Literature
In the existing scholarly literature on the new ONGO Law, various authors have focused on the legal texts; their interpretation and implementation; and the direct, mostly restrictive, implications for foreign civic actors (Kang 2018;Shieh 2018;Sidel 2019;Pissler 2016). A large share of the literature has discussed the ONGO Law as emblematic of a wider trend of shrinking or closing space for NGO activities worldwide, with a plethora of new or revised NGO regulations, media laws, anti-terror, anti-spy and other security laws restricting civil society's room to manoeuvre across the globe (Lang and Holbig 2018) and particularly targeting links between international NGOs and their local partners and funding recipients. In this context, the global dynamics of diffusion and learning between China, Russia and other authoritarian regimes struggling with the spectre of pro-democratic "colour revolutions" have attracted particular attention (Hall and Ambrosio 2017;Koesel and Bunce 2013;Lang 2018).
More recently, however, comparative politics scholars have argued that in various authoritarian and hybrid regimes "the space for civil society may be changing or shifting rather than just closing" (Toepler et al. 2020, 650). Thus, the room to manoeuvre may be increasingly limited for more outspoken, claim-making NGOs, while it is expanding for service-oriented non-profits in many parts of the world. "Loyal" NGOs willing to subscribe to the regime's cultural hegemony, align with norms and ideas of official discourse and act as vehicles for nation-building and modernisation efforts may even be offered privileged status and resources, thus contributing to the regime's stability and legitimacy (Toepler et al. 2020, 653-657).
This nuanced perspective on multiple patterns of interaction between INGOs, domestic non-profits and the relevant (party-)state authorities at the central and local levels builds on China scholars' previous work. To characterise the ambiguous, paradoxical nature of state-civil society relations, they have developed insightful conceptualisations such as "embedded autonomy" (Ho and Edmonds 2008), "graduated control" (Kang and Han 2008), "contingent symbiosis" (Spires 2011), "self-limiting organizations and codependent state-society relations" (Hildebrandt 2013, 1) and "consultative authoritarianism" (Teets 2013). Departing from this rich scholarship, our article investigates the longer-term dynamics underlying these interactions and gauges the future prospects for international civil society work in and with China.

Grasping the Dynamics of Institutional Change: Rules-in-Form Versus Rulesin-Use
Over the past few years, several scholars have employed various strains of institutionalist theory in studying China's NGO sector (Hsu and Hasmath 2013;Tam and Hasmath 2015). Gåsemyr (2017, 87) has conceptualised the various proactive strategies Chinese NGOs have used to navigate the multiple opportunities and risks, including strategies to "circumvent formal restrictions and broker functional arrangements that work within China's restrictive but relatively flexible institutional settings." In an insightful application of historical constructivism, a variant of institutionalist theories, Howell (2019) has analysed gradual institutional changes in the regulatory framework for social organisations and the state's efforts to craft a civic welfare infrastructure during the Hu andWen administration (2002-2012). The strength of this approach is its conceptualisation of path dependencies, critical junctures and the subtle nuances of gradual institutional change -elaborated through notions such as institutional layering, drift, conversion, and the like, observable over longer periods (and extended by Howell to include the new notions of "bounded adjustment" and "rule creation"). For this study, however, these variants of institutionalism not only appear too fine-grained conceptually to identify a manageable set of scenarios, but also tend to blur the dynamic interaction between formal-legal acts and informal behaviour which we believe can be captured in more parsimonious ways.
To do so, this article applies a perspective inspired by institutionalist economics. While legal scholars distinguish between "law-in-book" and "law-in-action," or between "written" and "unwritten rules," Ostrom (2005; makes a similar distinction between "rules-in-form" and "rules-in-use" (see also Schmid 2004). Rules are defined here as " . . . shared understandings among those involved that refer to enforced prescriptions about what actions . . . are required, prohibited, or permitted. All rules are the result of implicit or explicit efforts to achieve order and predictability among humans" (Ostrom 2011, 17). This definition includes formalised laws and regulations (but of course goes far beyond legislative acts) and makes it possible to capture the whole range of participants' behaviour, from implementation of a law both in the letter and the spirit; to partial and selective implementation; and to shirking, circumventing and outright ignorance of the law (Cole 2017;Ostrom 2008).
In this framework, rules-in-use are "normative understandings about what a participant . . . must, must not, or may do in a particular situation, backed by at least a minimal sanctioning ability for non-compliance . . . When these normative instructions are merely written in administrative procedures, legislation, or a contract and not known by the participants or enforced by them or others, they are considered rules-in-form" (Hess and Ostrom 2007, 50). According to Ostrom, in systems governed by the rule of law, rules-inform will mostly translate into and be consistent with rules-in-use. In other systems there "may be central laws and considerable efforts made to enforce them, but individuals attempt to evade rather than obey the law" (Ostrom 2011, 18).
Revisiting Ostrom's framework, Cole (2017) has offered a simple typology of relations between formal legal rules and rules-in-use, suggesting that inconsistencies between rulesin-form and rules-in-use are common even in systems with a strong rule of law. He distinguishes between configurations where legal rules are "so clear and controlling . . . that they require virtually no interpretation or conversion" into rules-in-use (Type 1); where legal rules that "could be" rules-in-use are not coextensive with the rules-in-use because they are "publicly known not to be strictly enforced" and where prevalent social norms exist that, in effect, translate the formal legal rule into different rules-in-use (Type 2); and, finally, where legal rules play "no significant role in the organisation of social behaviour because they simply do not affect social interactions," sometimes leading to an "ever-increasing reduction in the relevance of the formal rule" (Type 3) (Cole 2017, 839-842).
In other words, in systems with either strong or weak rule of law, a distinction can be made between types of institutional change where new laws, or rules-in-form, are largely ignored and have no lasting effect on the existing rules-in-use ("no change," Cole's Type 3); where new rules-in-form fully and consistently translate into new rules-in-use ("rules-inform become rules-in-use," Cole's Type 1); and where new laws are partly evaded, shirked or circumvented by actors at different administrative levels and in different regions and localities ("rules-in-use partly deviate from rules-in-form", Cole's Type 2).
In addition to these various types of institutional change resulting from the enactment of a new law, however, another outcome could be labelled "changing the rules of the game." Inspired by Greif and Laitin's (2004) theory of endogenous institutional change, one can hypothesise an outcome where a new rule-in-form (that is, a new law) leads to an endogenous reconfiguration of the existing rules-in-use, or where the rules of the game are changing while the game is being played. 2 While this is the most abstract type of institutional change, it is crucial to putting currently observable developments into a longer-term perspective. For brevity's sake, we label this type of institutional change an incremental endogenous change in the rules of the game.

Four Scenarios: No Change, Restraining, Recalibrating, Reorienting
From the different types of gradual institutional change derived from Ostrom's work and other strains of institutional economics, we can now develop theory-based scenarios for how the ONGO Law as a new rule-in-form will affect the institutional context of international civil society activities in the longer run. Based on the possible relations between rules-in-form and rules-in-use outlined above, the following four scenarios are conceivable (see Table 1): • Scenario 1: "No change." This scenario emerges from the hypothetical assumption that new rules-in-form are largely ignored and thus have no lasting effect on the existing rules-in-use. As we see in the empirical analysis below, while informal practices of ignoring the law were widespread in the 1990s and early 2000s, this scenario has lost its relevance in the Xi Jinping era and appears unlikely to gain prominence again anytime soon. • Scenario 2: "Restraining." A second scenario can be derived from the assumption of a rigorous implementation of the new ONGO law both in letter and spirit across the board (rules-in-form becoming rules-in-use). This would imply severe restraints for INGOs and their domestic partners in China. This scenario has received the most attention in the international "shrinking space" debate. We refer to it as the "restraining" scenario. • Scenario 3: "Recalibrating." A third scenario can be deduced from the "rules-in-use partly deviate from rules-in-form" type of institutional change. This refers to partial and inconsistent implementation of the new law, depending on the diverging motivations of bureaucratic actors in different sectors and at lower administrative levels as well as the positions and behaviours of INGOs and their Chinese non-profit partners, which are simultaneously shaped by the larger domestic regulatory environment. This scenario implies incremental institutional change embodied in gradual adaptation by some INGOs and partial crowding-out of others unable to adapt or unwilling to compromise their agenda, along with their potential partial replacement with domestic actors. In anticipation of the possibility of new configurations between INGOs and domestic NGOs, we refer to this as the "recalibrating" scenario. • Scenario 4: "Reorienting." Finally, a fourth scenario can be derived from the last type of incremental institutional change, conceptualised above as endogenous change in the rules of the game. In our case, we expect that the ONGO Law as a new rule-in- form directed at globally operating civil society organisations (CSOs) is but one element of a broader norm-shaping effort by the party-state to reconfigure existing rules-in-use in the transnational non-profit realm. This scenario is plausible to the extent that we can identify international actors and their domestic counterparts who are gradually "reorienting" their anticipations, organisational behaviour and strategic choices towards the "new game in town," resulting in a gradual alignment of the rules-in-use with norms actively shaped by the party-state. This is labelled the "reorienting" scenario. 3 With this theoretically informed set of scenarios in mind, the research questions can be specified. What kind of institutional change has been observed in the non-profit realm since the announcement of the new ONGO Law in late 2014, and particularly since its enactment in January 2017, and are there indications of rules-in-form only, of (partially new) rules-in-use, or even of endogenous changes in the national and international sets of rules-of-the-game? Which driving forces behind these different types of change can we identify, and what do these driving forces tell us about each scenario's consistency and plausibility? Last, what are the longer-term implications of the gradual institutional change observed in China for international civil society and how is the likelihood of the four scenarios' future occurrence affected by external scope conditions? After discussing these questions separately for the four scenarios in the following sections, the plausibility and likelihood of each scenario is assessed in the conclusion.

No Change: Maintaining the Status Quo through Reliance on Informal Rules
Over the more than two decades of China's reform period, the existence of a myriad of unregistered grassroots NGOs indicated that most civic actors, often colluding with local party-state actors who welcomed their operations on the ground, were able to ignore the official laws and regulations. Estimates for the Hu-Wen administration (2002-2012 varied between one and 2.7 million unregistered domestic NGOs and between 1,000 and 3,000 international NGOs and foundations operating in China without registration (Holbig and Bälz 2018, 104). The massive divergence between rules-in-form and rules-inuse was possible due to a pragmatic attitude on all sides of tacitly acknowledging operations in legal grey zones. Chinese analysts explained this paradox as stemming from compliance with an unspoken rule of "no recognition, no banning, no intervention," a situation also mocked as you fa bu yi (literally, "laws exist but are not applied") (Deng 2010, 201). With this previous situation in mind, a no change scenario would indeed be conceivable. Party-state as well as foreign and domestic civic actors at the local levels would either simply ignore the new ONGO Law or find ways to continuously shirk the new stipulations for registration and public security monitoring, relying instead on trusted informal relationships. However, interviews and empirical insights since 2017 (notably E01, E04, E05, Int 01, Int11, Int12) lead us to discard such a scenario, because significant, irreversible changes in the institutional set-up and behaviour of all actors involved took place during the law-making process and have continued since its enactment.
An early example of the intimidating effect of the new law's announcement in late 2014 is the case of Nanfeiyan, a grassroots NGO established in Guangdong Province in 2012 to protect migrant worker rights. While it started out as an unregistered NGO due to the sensitive nature of the related labour issues, the local government asked its founder to register the NGO as a private non-enterprise unit, which allowed the NGO to legally receive government funding. Nanfeiyan contracted municipal grants to provide afterschool care for migrant children and shelters for homeless migrants in 2012, 2013 and 2014. To increase funding, it also tapped into non-governmental sources, including international sponsors such as Oxfam (Int01). Directly after the announcement of the ONGO Law in late 2014, however, the local public security bureau stepped up its monitoring of Nanfeiyan's ties to foreign funders and local party-government collaborators started to withdraw their support. They signalled that they would revoke Nanfeiyan's registration and rescind funding on administrative grounds. Over the course of 2015, the NGO was driven underground and its founder arrested on charges of embezzlement, only to be released on bail four months later in April 2016 (Yuen 2018).
While this episode might represent a rather dramatic case, it suffices as a counterfactual argument to corroborate the obsolescence of the "no change" scenario early on, in mere anticipation of the enactment of the ONGO Law. More recent publications suggest that local party-state actors' pre-emptive caution in dealing with internationally funded domestic NGOs, as evident in the Nanfeiyan case, has evolved into a broader behavioural pattern. For example, in an insightful study of CSOs' funding practices, Fulda and Hsu (2020) have proposed a "resource mobilization cycle" that has allowed some entities to leverage cultural, economic, symbolic and social capital, thereby enabling them to secure funding from various sources, including from foreign, government and private sources. The authors, however, also point to the trade-off between foreign and domestic government funding due to the conflicting pressures accompanying domestic and foreign funders' agendas. In light of the ONGO Law, they expect a significant reduction in Chinese CSOs' foreign funding, which will pressure them to raise more economic capital from domestic sources in future (Fulda and Hsu 2020, 75-76).
In a study of institutionalised forms of NGO co-optation by local governments, Chinese analysts have even identified a "red line" regarding the perceived sensitivity of tapping into foreign funding in recent years. " . . . [T]he connections with the foreign groups and funding sources," they argue, "are increasingly restricted under the Xi Jinping . . . administration. The red line here is quite clear and the NGOs do not need to reach out to the supervision agencies to know that they cannot cross the line" (Li and Wang 2020, 719). Exceptions to this rule might exist -for example, in the environmental realm, where local alliance-building with foreign donors remains politically feasible (Zhang, Bradtke and Halvey 2020, 790) -but overall, INGOs appear to have been perceived and treated by domestic actors with growing anxiety in the wake of the ONGO Law. The "no change" scenario thus seems increasingly implausible in both the short and longer term.

Restraining: Consistent Implementation Leading to Shrinking Spaces across the Board
The securitisation trend can be traced back to the latter half of the 2000s when the working environment for foreign civil society actors started becoming visibly more difficult. Chinese policy debates on NGOs used to be roughly divided between proponents of a nonconfrontational, service-oriented civil society with input from Western countries acknowledged as useful, and critics portraying INGOs as a potentially subversive US foreign-policy tool. The US government's perceived support for grassroots civil society and oppositional social movements was viewed by the latter as instrumental in the making of the Central Asian Colour Revolutions and the Arab Spring revolts. Since Xi Jinping's ascent to paramount leadership, crackdowns on previously tolerated advocacy activities, such as by independent unions, women's rights groups and human rights lawyers, and the banning of domestic debates on "civil society" (gongmin shehui), denounced as a "political tool . . . adopted by Western anti-China forces," have intensified the impression that authorities now see INGOs primarily as a security threat (ChinaFile 2013).
Accordingly, the decisive push for the ONGO Law, a draft of which was first circulated for public comment in mid-2015, came from the National Security Commission, one among various high-powered organs created at the CCP headquarters in 2013 and chaired personally by Xi (Shieh 2018, 5). This was despite the fact that more moderate pilot regulations on INGO management had been experimented with at the provincial level. The origin of the national law in top security circles explains why regulatory competence for INGOs was transferred from the Ministry of Civil Affairs, responsible for supervising the domestic third sector, to the MPS and lower-level Public Security Bureaus. With the enactment of the ONGO Law, international non-profit organisations are now formally overseen by police authorities -clearly indicating the securitisation of China's policy towards international civil society.
The new ONGO Law's Article 2 leaves foreign "foundations, social groups, think tanks and other non-profit, nongovernmental social organizations" two avenues for carrying out legal activities: either by seeking Public Security Bureau approval for a registered representative office or by asking a Chinese partner organisation to file records for temporary activities of no more than one year (Articles 9-11). In addition to imposing onerous reporting requirements, set out in Article 12, Articles 39-43 give public security extensive "oversight and supervision" competencies, the law also subjects INGOs to the dual registration system, which requires a Chinese sponsor organisation or professional supervisory unit (yewu zhuguan danwei, PSU) that effectively vouches for all activities carried out by an INGO in China. This system has been subject to significant debate since the 1990s and has been partially abolished for domestic social organisations. Further restrictive measures in the ONGO Law, such as the prohibition on engaging in or funding "political activities" of any kind (Article 5) or the possibility of detaining ONGO staff for supporting advocacy work -worded in Article 47 as "inciting resistance to laws and regulations" or "spreading rumours," are best understood as a formalisation of previously informal state powers to control and potentially intimidate foreign and Chinese NGO staff alike (see Pissler 2016;Lang and Holbig 2018).
Besides reflecting this larger securitisation rationale, the ONGO Law also exemplifies a new approach to societal control developed by the party-state in the later years of Hu Jintao's leadership and systematically institutionalised under Xi Jinping. The official substitution of the previous technocratic approach to "social management" (shehui guanli) with a "social governance" (shehui zhili) approach was framed within the broader rhetoric of Xi's claim to be modernising state governance. The introduction of this governance approach was accompanied by the formalisation of an increasingly rigorous hierarchical system of societal control flowing down from the paramount leadership's ambitions to "govern the country in accordance with the law" (yifa zhiguo) while centralising decision-making through "top-down design" (dingceng sheji) (Holbig 2018, 191). At least from the perspective of formal institutions, the ONGO Law, together with the regime's larger efforts to restructure party and state organs since 2013, clearly exhibits the leadership's explicit claims to top-down societal control.
The most significant recent institutional changes demonstrate that the political surveillance of domestic and foreign non-profit organisations in China has been stepped up through the massive expansion of the CCP into grassroots society. In September 2015, a CCP document pushed for the intensification of party-building work in social organisations (CCP Central Office 2015). After a slow start, many domestic non-profits decided to install party cells or liaise with party cadres to demonstrate their loyalty to the ruling CCP (Liu and Van de Walle 2020, 741-742).
In March 2018, the 13th National People's Congress adopted the fifth constitutional amendment to the PRC's state constitution, which enshrined "Xi Jinping's Thought on Socialism with Chinese Characteristics for a New Era." Simultaneously, the CCP published a 40-page "Plan on Deepening Reform of Party and State Organs," which laid out a detailed scheme of structural reforms, including the creation of a new, powerful National Supervision Commission (guojia jiancha weiyuanhui). Practically speaking, the new commission was designed as the long arm of the CCP Central Committee's Discipline Inspection Commission. Under the auspices of the National Supervision Commission and its local counterparts, the nation-wide anti-corruption campaign launched by Xi in 2013, which had so far been directed mainly at party members, was formally expanded to target all public servants from the central level down to the county level (Holbig 2018). Also, the 40-page document demanded that all mass organisations revitalise their original -Maoist -"role as bridges and transmission belts between the party-state and the people's masses" (dang he zhengfu lianxi renmin qunzhong de qiaoliang niudai zuoyong) (CCP Central Committee 2018).
Taken together, these recent efforts by the party-state to institutionalise tighter topdown control of public servants and societal actors at large have not formally and immediately affected the activities of INGOs in China. However, they have tended to enhance the restraining effect of the ONGO Law by subjecting the previously widespread practices of informal collaboration between party-state cadres and civic actors at local levels to suspicion of illicit collusion (see "recalibration" scenario below).
The previous rules-in-use of informal collaboration or what Spires (2011) calls "contingent symbiosis" at the local level are now being systematically targeted by party-state measures in the "New Era." The disciplinary campaign means local administrative and public service units now must ponder potential charges of collusive behaviour when considering maintaining or entering into new co-operation projects involving grassroots NGOs, particularly those with foreign ties. Since 2018, official propaganda material produced for domestic consumption has repeatedly mentioned overseas NGOs as a potential security risk. Official propaganda videos have used (fake) ONGO employees as an example to warn against "foreign spies," while education departments have stressed the risk of "foreign ideological infiltration" in the societal sphere (ChinaFile 2018). This contrasts starkly with the government's external communication surrounding the ONGO Law, which has maintained a conciliatory tone throughout.
The restraining effects of the formal ONGO Law have thus been reinforced by the massive restructuring of party-state organs and the institutionalisation of anti-corruption campaigns, resulting in gradual institutional change that has substantially modified previous rules-in-use. The actors involved appear to perceive the new legal environment not so much as increasing the transparency and reliability of rules, but rather as creating more uncertainty (Int25, E01). To put it mildly, party-state cadres and civic actors at the local level now have significantly more reason to worry about acting as sponsors, accepting foreign funding, or tacitly acknowledging unregistered activities (Holbig and Bälz 2018, 98-99). It is in this sense that the ONGO Law has become an effective element in the toolbox of party-state control over (civil) society (Jia 2018), and that new rules-inform have been moulded into dominant rules-in-use during the implementation of the ONGO Law.
Against this backdrop, the "restraining" scenario appears plausible to the extent that the forces within the CCP leadership driving the securitisation of societal governance under Xi Jinping's reign fully prevail over competing bureaucratic interests (thereby also undermining the achievements of other central-level policy goals). In light of escalating Sino-USA tensions, the draconian enforcement of the National Security Law in Hong Kong and the repercussions of the Covid-19 pandemic throughout 2020, this scenario dominated by security concerns and distrust has certainly become more likely.

Recalibrating: Partial Implementation, Crowding-Out of INGOs by Chinese Social Organisations
The "restraining" scenario, which implies that national security hardliners in the central leadership will fully prevail over competing bureaucratic interests, is not the only plausible scenario. Bureaucratic wrangles, diverging sectoral agendas and local vested interests remain a core feature of Chinese policymaking, despite Xi's sweeping efforts to streamline the party-state bureaucracy from the top (see Ma 2020;Gåsemyr 2017). From an institutionalist perspective, an alternative scenario assuming enduring partial deviation between rules-in-use and rules-in-form is supported by empirical evidence that, as in other policy fields, the party-state's approach to non-profit organisations including INGOs continues to be ambiguous and driven by diverging domestic and transnational policy agendas. The resulting inconsistent implementation of the ONGO Law, if continued in the longer term, will amount to a recalibration rather than a flat-out restriction of spaces for (I)NGO activities.
Starting from the central-level master document for policymaking, the 13th Five-Year Plan (2016-2020), the overarching goals defined for different policy areas have diverging, sometimes contradictory implications for civil society policy. From a central government perspective, the need to restrain foreign actors' influence on society must be reconciled with the increased use of private resources and third-sector expertise to achieve crucial development goals (E06). Thus, the 13th Five-Year Plan reiterates the "strong support to the development of professional social work and charity" and pledges to "mobilise nongovernmental actors to provide social relief, mutual aid, and volunteer services" (NDRC 2016, Ch. 64/3). Among the development goals to be achieved with the support of "social organisations" or "non-governmental actors" are poverty eradication, improvement of healthcare coverage, social insurance, elderly care and rural education (NDRC 2016, Ch. 64/3). This builds on the State Council's (2014) "Guiding Opinions on the Healthy Development of Charities," which clearly spelled out the government's resolve to tap into private resources to support its own development agenda, while also strengthening supervision and confining charitable activities to more narrowly confined issue areas.
While INGOs have played an important role in these fields for many decades (Int08, Int15, E01, E03, E05), only domestic social organisations, which are supposedly less likely to pursue their own agendas and more easily controlled by the party-state (Int02), are mentioned. This supports the idea that untrustworthy and allegedly subversive INGOs, denounced as "anti-China forces" (fan Hua shili) in Document No. 9 (ChinaFile 2013), are meant to be partially and gradually crowded out by homegrown counterparts, in so far as the latter can effectively assume similarly "useful" roles. Such a recalibration of civil society policy, including a deliberate preference for domestic over foreign non-profits, is substantiated by contrasting the ONGO Law with another rule-in-form passed almost simultaneously to regulate domestic non-profits: the Charity Law (cishan fa).
Both the ONGO Law and the Charity Law are part of Xi Jinping's new governance agenda of "governing the country in accordance with the law" (yifa zhiguo), which aims to formalise the rules for sectors previously governed by administrative regulations that were more or less ignored in practice (Trevaskes 2018, 351-356). Indeed, notwithstanding earlier government endorsements of domestic charities, notably the State Council's 2004 "Regulations on Foundation Management," the evolution of China's charity sector was largely left to informal negotiations between public officials and private entrepreneurs. This institutional set-up did not, however, achieve the desired outcomes from a civil affairs perspective, nor did it prevent the abuse of supposedly non-profit organisations for public-private collusion or private enrichment (State Council 2014). When the highly problematic rules-in-use of charity development came to light through a series of scandals involving fraud and extravagance in mostly state-affiliated charities around 2011 (Shpakovskaya 2017), third-sector scholars and practitioners reinforced their push for institutional change, including clearer, more transparent formal rules for the registration and operation of charitable organisations (Int02, Int03, Int22). In response to these calls, the eventual adoption of the Charity Law in March 2016 was intended to restore public trust in charity and shore up the low levels of private donations (see Lang 2018, 165-170), albeit under tight security constraints imposed in the broader political context of tightened top-down control and general mistrust of private initiatives (Kang 2018).
Thus, both laws broadly follow an agenda of formalising rules and thus curtailing the perceived abuses observed in informal relations between state and non-state actors. But the exclusion of foreign-based entities from the Charity Law's remit along with the adoption of a separate law for overseas NGOs has created a rigid separation between domestic and foreign non-profit organisations and even instituted supervision by different ministries -the Civil Affairs Ministry for Chinese organisations and the Public Security Ministry for foreign non-profits. 4 Notwithstanding the transfer of formal responsibility for managing overseas NGOs to the Public Security Ministry's apparatus, other ministries and local-level governments are much more directly involved with and experienced with domestic and international nonprofit organisations. Indeed, many local government bodies used to actively solicit foreign funding in support of their own policies, including through the creation of "government-organised NGOs" (GONGOs) specifically for that purpose (see Spires 2011;Jia 2016). Non-profits maximised their own agency by becoming embedded in the local state and often had good reasons to remain unregistered (Yuen 2018, 408-413).
As part of the widespread informal practice, local cadres benefitted from the nonregistration of non-profits, since this meant external funding to these grassroots groups had to be channelled through their hands (Hildebrandt 2011). These symbiotic or collusive triangular relationships between local government agencies, Chinese GONGOs and INGOs prevailed during the 1990s and 2000s (see, for example, Hsu and Hasmath 2013;Hsu and Teets 2016).
Over time, however, central-level policymakers worried about local authorities' distorted incentives in regulating INGOs. In line with the sweeping anti-corruption and disciplinary campaign of President Xi (2013), who explicitly pledged to "no longer tolerate any form of 'top-down policy measures and bottom-up counter-measures'" (jue bu yunxu "shang you zhengce, xia you duice"), disciplinary measures for suspected regulatory laxness on the part of local authorities were stepped up significantly (Kang 2018, 799-800). However, this more centralised and discipline-focused governance approach also directly clashed with the many experimental and innovative approaches of state-NGO collaboration practiced at lower levels thus accentuating intra-bureaucratic conflicts of interest (see Yan, Lin, andRen 2017, Yuen 2018).
That the draft ONGO Law emerged from security circles which understand their civil society-related tasks primarily as preventing "Western infiltration" explains why its text is heavily control-focused (ChinaFile 2013). Other bureaucratic interests did, however, come into play between the initial draft and its -postponed -adoption in April 2016. Compared with the draft initially circulated for public comments in mid-2015, several moderating changes were incorporated in the final legal text (Jia 2016;Shieh 2018). For instance, the prohibition on overseas NGOs accepting funds from within China in the second draft represented the clearest sign that INGOs were meant to be replaced with domestic fundraising non-profits instead of competing with them for Chinese donors' charitable contributions. In the final version, this incisive stipulation was reduced in Article 21(3) to a prohibition on soliciting donations (jinxing mujuan), which most legal scholars have interpreted to mean that merely accepting (formally non-solicited) donations from China is still permitted (Pissler 2016, 121). In practice, INGOs' experiences suggest that provincial and local authorities' interpretations range from explicitly prohibiting all sorts of fundraising -probably due to limited knowledge of the law's final revisions -to allowing and even encouraging INGOs to accept officially unsolicited funds to be used in their administrative remit (Int15, E05).
Similarly, the initial prohibition on INGOs opening more than one representative office was rescinded in the final draft (Pissler 2016, 121). This meant that many larger INGOs could still operate several provincial bureaus in addition to their main office, which is typically in Beijing (164 offices) or Shanghai (113). This possibility is crucial for service-oriented INGOs strongly embedded in less developed areas in China's interior, which rely on permanent staff on the ground to ensure both the functioning of the programmes and the cultivation of good relations with local officials essential to their survival (Int05, Int13, Int15). These amendments to an initially more restrictive draft law can be interpreted as the result of both foreign pressure (that is, China being sensitive to diplomatic pressure to some extent) and lower-level bureaucratic interests, notably from more peripheral provinces where cash-strapped officials have much higher stakes in attracting foreign funding and expertise (Zhang 2015).
Judging from the early implementation phase, the ONGO Law has not led to a mass exodus of INGOs from mainland China. Instead, the official registration figures (see Table 2), which are now regularly published and updated on the "Overseas NGO Service Platform" (MPS 2020) and which have indeed greatly enhanced transparency regarding the presence of INGOs in China, show evidence of what we term a recalibration of INGOs' roles.
As Figure 1 further demonstrates, Batke and Hang's (2018) early observation that INGO activities were shifting towards fields of activity high up the government's domestic policy agenda is strongly confirmed by the most recent registration figures, which show that INGOs' main areas of work now very much align with those of homegrown Chinese charity organisations, with "education," "poverty alleviation," "youth" and "health" issues topping the list (see MPS 2020). Despite the absence of reliable data on INGO activities prior to the ONGO Law, extant research suggests that the focus areas were substantially different, and included financial support for Chinese grassroots  organisations and work in more political fields such as labour and gender issues and legal and governance reform (see Spires 2011;Xie 2011;Zi and Bullock 2014). Sectoral interests from those segments of the bureaucracy concerned with foreign relations and China's international image have also contributed to softening the law's restraining effects in the implementation phase. The absence of direct registration refusals by police authorities to date suggests that the exclusion of undesirable foreign organisations mainly takes place informally to limit foreign criticism. The overwhelming red tape involved in planning, documenting and reporting organisational activities in excruciating detail (Int19, Int21, Int25) and the need to find a well-meaning sponsor agency from a limited whitelist of PSUs (MPS 2019) allow for the silent exclusion of unwanted organisations that are simply unable to find a Chinese sponsor willing to vouch for their activities (E05, Int12, Int22). At the same time, the bureaucracy has demonstrated political flexibility and pragmatism in the law's early implementation phase (Jia 2018), as shown by the continuous ad hoc expansion of the PSU whitelist (MPS 2019; E01, E05).
Local governments tasked with poverty alleviation and social development, especially in China's poorer regions, also have vested interests in maintaining long-standing working relations with INGOs, especially if trusting personal relationships have been established over time (E04, Int05, Int15, Int25). Thus, representatives of INGOs with a longer track record of local-level social programmes reported only minor changes due to the ONGO Law in conversations in September 2018 and June 2019 -apart from tedious additional bureaucratic paperwork (Int09, Int10, Int15). The importance of such organisation-specific factors -rather than a systematic difference between regions -has also been highlighted in previous case study research on the ONGO Law's implementation (Li 2020). Through our theoretical lens, this means that in some cases informal rules-in-use may be continuing to work reasonably well for all parties involved (E06; Jia 2018).
An additional case in point for the ad hoc and informal nature of implementation is the possibility -not foreseen in the law but tolerated for a few well-established INGOsof establishing a local foundation with the central purpose of raising funds for work in China, since the same would be unfeasible for the INGO itself according to the letter of the ONGO Law (Int17, Int21). And there are still instances where smaller INGOs can work without registration for the time being with the tacit consent of local officials (Int07).
In sum, whereas the political environment for activities in sensitive fields and advocacy work more generally had already become less flexible in previous years, the law's oppressive potential has been scarcely used to date (Int02, Int13). Instead, the more serious problems affecting INGOs' work in China include dwindling foreign funding for activities in China (Int09, Int17) and the retreat of Chinese partner organisations -due to their own political difficulties, to the red tape involved in accepting foreign funding, or simply out of caution (E05, E02, E01). Also, since 2019, the political escalation in Hong Kong has derailed several INGOs' fundraising strategies and made them the target of criticism from democracy activists for being complicit in Chinese authoritarianism (E05, Int15, Int16, Int24).
Despite the CCP's sustained crackdown on political dissent and its aggressive quest for social control over the past decade, China's social work (shehui gongzuo) and charity (gongyi cishan) sectors have been reinvigorated through the proliferation of related study curricula launched across the country and the government's booming purchases of social services from NGOs (Wang and Snape 2018;Ma 2020). Prior to the introduction of the ONGO Law, many of these social policy tasks were supported by INGOs through funding, capacity-building and advice based on international models and experiences (E01, E02, E03, Int16, Int21, Int22, Int24). Now, the spaces for doing such work informally have significantly diminished, while the ONGO Law has put onerous burdens on INGOs willing to support Chinese grassroots NGOs. But even if foreign donors are to be gradually replaced by domestic charities, "volunteer services" (zhiyuan fuwu) and "social work entities" (shehui gongzuo jigou), this process will take time. Private foundations in particular still lack the capacity and sometimes the willingness to establish reliable donor-beneficiary relationships with formerly foreign-supported NGOs (Int06; see also Shieh 2017). At the same time, foreign donors have significantly contributed to the development and professionalisation of the Chinese non-profit sector. INGOs are now prioritising capacity-building support for would-be Chinese grant-making charities, including direct staff training (Int10, Int13, E02, Int25) and the sponsorship of professional associations such as the China Global Philanthropy Institute or the China Foundation Forum. 5 In the longer run, foreign funders may thus contribute to making themselves redundant in the eyes of Chinese policymakers -not unlike the trend observed today in many for-profit industries.

Reorienting: Towards a "New Game in Town" for Global Civil Society
While the "recalibrating" scenario assumes the patchy and selective implementation of the ONGO Law due to fragmented bureaucratic interests, the final scenario focuses on the broader international influences and motivations behind Chinese policies towards INGOs. Its core tenet is a more fundamental, gradual reconfiguration of institutions and practices in the non-profit realm, whereby the party-state's regulatory ambitions have the potential to alter civil society norms and practices on a global level, leading to an endogenous change in the rules of the game.
The "reorienting" scenario is supported by the observation that, in the course of recent institutional changes, the main purpose of INGOs' presence in China in the eyes of partystate actors has shifted: whereas INGOs' direct financial and technical contributions to national socio-economic development goals have been valued, especially by local government actors, for decades (see "recalibrating" scenario), since the proclamation of the ambitious Belt and Road Initiative (BRI) central-level party-state actors have started viewing selected INGOs as foreign partners in international people-to-people co-operation (minjian hezuo) under China's own framework for international co-operation (Philanthropy Times 2018).
While the Western concept of civil society (gongmin shehui) has mostly disappeared due to its alleged abuse by "anti-China forces" (ChinaFile 2013), Xi Jinping's (2017) declared will to "strengthen people-to-people co-operation" has sparked numerous calls for civic organisations (minjian zuzhi) to "go out" (zouchuqu) and intensify relations with international counterparts since 2017 (Mu 2017). Henceforth, activities ranging from NGO exchanges and youth events to cinema and pop culture are all subsumed under the BRI's fifth pillar, with the poetic Chinese heading "letting the people's hearts communicate" (minxin xiangtong). 6 This has increased the pressure on INGOs to align with the party-state's global ambitions and reorient their work towards activities supplementary to governmental foreign policy agendas, such as non-political cultural or youth exchanges. This new party-state agenda, dominated more by diplomatic than by social policy concerns, has important implications for the kinds of INGO activities and behaviour that will be condoned and encouraged in the future.
From this perspective, the ONGO Law as a new rule-in-form is only the visible tip of the iceberg of a much broader, incremental reconfiguration for foreign actors operating in and with China. Whereas access to the PRC has always meant respecting the explicit and implicit restrictions on civil society work inherent to the authoritarian one-party system -a trade-off pointedly described by Noakes (2018) as the "advocacy trap" -it now increasingly means subscribing to a specific understanding of civic organisations as "constructive forces" (jianshexing liliang) and protagonists of "people-to-people connectivity" (Xinhua, May 15, 2017; see also Hu and Wang 2014). This implies a tacit acceptance of the CCP leadership's downplaying of civil society's relevance as an independent -and potentially power-critical -realm of society. The ONGO Law has served to intimidate and better control organisations unlikely to subscribe to such a paradiplomatic agenda, yet it has also required all others to reorient their China programmes by looking for sector-specific PSUs and designing new programmes likely to be accepted by supervisory authorities (E01, E05, Int13, Int17). This formal push towards reorientation has been complemented since 2017 by formal and informal Chinese initiatives to create new spaces for NGO activities under the BRI umbrella. Most notably, the "List of Deliverables" from the high-level Belt and Road Forum held in May 2017 in Beijingcertainly the most authoritative document laying out the BRI agenda -notes that: China NGO Network for International Exchanges and over 80 Chinese NGOs jointly launched the Chinese Social Organizations' Action Plan for Stronger People-to-People Connectivity along the Belt andRoad (2017-2020). China NGO Network for International Exchanges and over 150 civil organizations jointly set up the Silk Road NGO Cooperation Network (Xinhua, May 15, 2017).
These announcements are essentially an implementation of the 13th Five-Year Plan's Chapter 51, entitled "Move Forward with the BRI," which specifically defined the issue areas for such politically welcome "international cooperation in the areas of education, science, technology, culture, sports, tourism, environmental protection, health care, and traditional Chinese medicine" and vowed to "create mechanisms for official and nongovernmental cultural exchanges that involve the participation of multiple parties" (NDRC 2016, 51/3).
Under the "reorienting" scenario, the domestic security concerns about INGOs are counterbalanced by diplomatic concerns regarding President Xi's flagship foreign policy project, the lasting success of which relies on the cultivation of China's international image and reputation (Callahan 2016, Deng 2018. Major setbacks to key BRI projects, partly caused by anti-Chinese demonstrations and political campaigns (ranging from Vietnam, Myanmar and Malaysia to Kenya and Ecuador), are likely to further strengthen these concerns about China's image abroad. Considering these diplomatic challenges, foreign policy elites are notably interested in promoting a harmonious image of congruence and mutual complementarity between China's BRI and the UN Sustainable Development Goals (China Daily, March 8, 2017).
The contrast between the shockwave that the ONGO Law sent through US and Western European civil society communities and the absence of reactions from the rest of the world also shows how much the effects of this rule-in-form are mediated by other, not immediately apparent environmental factors. Indeed, the quasi-absence of INGO offices or even temporary activity filings from key BRI target countries in the MPS's official ONGO registration statistics suggests that informally, these activities are not even expected to be part of this securitised framework. Instead, they are viewed as part of the CCP's "South-South Co-operation" (nannan hezuo) framework, which requires "constructive" NGOs from BRI countries to organise officially sanctioned regional NGO fora and advance "people-to-people connectivity" through youth exchanges, non-governmental delegation visits, and scholarship programmes (see Hu and Wang 2014;Mu 2017).
Non-governmental exchanges, from this perspective, are a welcome complement to official diplomacy, and another way of "enabling China's voice to be heard by the world" (Tsai 2017). Chinese and foreign NGOs or social organisations are meant to become instrumental in "advancing the construction of the Belt and Road" (Hu and Wang 2014, 45-46). This mission is made very explicit in new venues created specifically for the purpose of "civic" (minjian) co-operation under the BRI: Thus, the first, high-level Silk Road NGO Cooperation Network Forum organised in Beijing in November 2017 was entitled "Jointly Building the Belt and Road and the Community of Shared Future for Mankind -the Mission of Civic Organisations" (see Mu 2017). The fact that this conference, which openly promoted the internationalisation of Chinese NGOs under the guidance of the Communist Party, was supported and co-sponsored by several Western foundations and gave rise to an "NGO network" with 153 Chinese social organisations and 199 members from 71 countries provides a blueprint for Sino-foreign engagement in accordance with a new model of international civil society with Chinese characteristics. Figure 2 illustrates the dramatic shift in the geographic focus of Chinese "civil society" co-operation: most SilkRoad NGO Network (Sironet)  Meanwhile, Chinese foundations are actively encouraged to learn from US foundations' (supposed) role in public diplomacy and look for international partners, especially in Western countries, to advance their own internationalisation (see Lang 2018, 167-169 andLang 2019, 35-40). Chinese scholars like Tsai (2017, 45) call upon foundations to "perform their innovative duties in BRI construction." The interviewees for this study spontaneously referred to the government's "going out" policy for social organisations as an incentive to seek out partnerships with European and US non-profits (Int03, Int06, E02, E03; see also Mu 2017). Capacity-building co-operation for Chinese and foreign non-profit leaders is equally encouraged, as long as the focus is clearly on strengthening the administrative capacities of China's non-profit sector rather than on sensitive issues such as advocacy and grassroots civil society support (E02).
While societal exchanges with Western countries are regarded as politically sensitive and are increasingly scrutinised, orchestrated exchanges between social organisations representing their people in explicit support of governmental friendship agendas represent a new model for promoting the societal component of the BRI (Int23, E04), along with an underlying redefinition of what "societal participation" means in international relations. The "reorienting" scenario appears to be supported by the party-state's recent international ambitions to redefine civil society participation at the UN level. This has been done, for example, using Chinese NGOs to "convey China's good voice" (chuanbo Zhongguo hao shengyin) and to introduce CCP slogans such as the "Community of Shared Future for Mankind" (renlei mingyun gongtongti) into the non-governmental sphere and the UN lexicon (Int12, Int14, Int20; see Liu 2016).
Finally, another far-reaching tool of informal control repeatedly mentioned in interviews with INGO staff is the increasingly frequent police pressure exerted on INGO representatives in China over their colleagues' publications and public statements regarding Chinese influence in other parts of the world (Int09, Int13, Int20, E05). Two INGO leaders explicitly mentioned that their China offices received inquiries from their supervisory Public Security Bureaus regarding English-language publications put out by the INGOs' headquarters or sister organisations in other countries that were perceived as critical of China's presence and behaviour in developing countries (Int13, Int21). Both interviewees expressed related concerns about the safety of their Chinese staff. Moreover, it is now tacitly acknowledged in INGO circles that maintaining a physical presence in China comes with a need to tread more cautiously in terms of global advocacy and messaging, meaning that China cannot be openly criticised just like any other government (E05). Thus, the CCP has started to use its leverage over local INGO branches to promote self-censorship on China-related statements on a global scale.
An obvious question that arises from this scenario: Why would foreign non-profit organisations play along with these new rules of the game? Under the rules-in-use dominating the Chinese civil society playing field before the ONGO Law's enactment, most INGOs were keenly aware of the informal boundaries of their work in legal grey zones (E01, E03, Int 05, Int11). While accepting that open criticism of governmental action -a core feature of their advocacy work in other countries -was impossible, most of them justified their presence in China with their contributions to universal values, for example, women's and children's rights, animal welfare, LGBTQ issues or the fight on climate change (E05, Int07, Int19). A survey into INGOs' adaptive behaviour by Noakes and Teets (2020) also suggests that these large transnational organisations have long been adjusting their modes of operation to the specific restraints of China's political system, including both conscious, strategic adaptations (such as an organisational focus on collaboration with local governments and political risk management) and socialisation effects that have profoundly changed organisational identities.
Since 2017, however, the institutional environment has been changing far more profoundly than a look at the ONGO Law would suggest. Interviews with leaders of Chinese GONGOs leave no doubt that INGOs looking for continued access and cooperation perspectives should become part of networks such as the Silk Road NGO Cooperation Network and contribute to the construction of the BRI's social component (Int23). And indeed, at least the more service-oriented INGOs that have managed to register under the ONGO Law are increasingly complying with these informal rules. This is most visible in terms of negative norms, that is, INGOs' avoidance of ever-increasing taboo areas such as support for grassroots NGOs engaged in public advocacy (including labour groups, LGBTQ activist groups or NGOs providing legal support to victims of state actions) through both project activities and public statements; however, it is also increasingly evident in terms of positive norms related to the BRI's "people-to-people connectivity" pillar.
In contrast with the increasingly vociferous criticism of the CCP regime's human rights violations or its encroachment on Hong Kong's freedoms in Western media and from rights-based INGOs, many established INGOs, especially those in the humanitarian field with a long tradition of service-orientation and partnership with governmental actors, are instead banking on an expansion of their activities in and with China over the coming decade (E03, E05, Int22). Among the strongest incentives to do so, interviewees regularly mentioned the sheer size of the country and the correspondingly large number of marginalised people in need of support, as well as the potential to enlist Chinese philanthropists as donors in times of dwindling Western development aid and philanthropic support (Int13, Int21).
From an institutional perspective, the "reorienting" scenario amounts to an incremental change in the rules-of-the-game steered from above but also shaped by Chinese social organisations and by those INGOs which prefer to continue playing the game rather than abandoning their China work altogether. The very meanings of civic engagement and civil society participation in international relations are being redefined in ways that encourage the participation of Chinese and foreign actors willing to play by a CCPdefined or, at least, a CCP-influenced, rulebook, while discouraging the involvement of confrontational advocacy groups. If these trends continue, and over-zealous national security officials do not totally undermine INGO activities in and with China, we can expect an expansion of joint-venture-like co-operation between Chinese and foreign non-profits on issues which are not only politically non-sensitive, but also further the Chinese government's broader foreign policy agenda and support the envisaged internationalisation of Chinese social organisations.

Conclusion
This article has demonstrated the usefulness of Ostrom's theory of institutional change in capturing the longer-term interaction between formal and informal dynamics of civil society regulation under authoritarianism. The scenario analysis derived from this theory has enabled us to view China's hotly debated Overseas NGO Law in a broader context and to assess its implications at the intersection of domestic and transnational developments. The analysis in this article has confirmed that the institutional environment for foreign and domestic civic actors in China has changed significantly over the past decade, a process that culminated but did not start or end with the ONGO Law's enactment in January 2017. As this investigation has shown, changes in formal and informal institutions can and should be analysed together, not only with a focus on new rules-in-form, but also taking into consideration the dynamic interaction between rules-in-form, rulesin-use and incremental reconfigurations of the rules of the game for the international third sector.
In addition to this contribution to the theoretical debate, this empirical research has identified competing trends and policy agendas within the party-state, which may explain why the CCP's current attitudes towards INGOs continue to be ambiguous, even after the formulation of a sweeping law intended to set clear standards for the whole sector. Considering that the timespan since the law's enactment has been too short for definitive conclusions from the institutionalist perspective adopted here, we have developed four scenarios regarding Chinese authorities' evolving approach to international civil society actors. While these scenarios serve as a heuristic tool for thinking about alternative future trajectories of gradual institutional change, their real-world occurrence may not be entirely mutually exclusive.
The "no change" scenario assumes that the rules-in-use will remain largely unaffected by the ONGO Law. As shown, this scenario is the only one to be discarded as there appears no way back to the "good old times" of informal co-operation.
The "restraining" scenario has proven plausible to the extent that the security apparatus will get its way and the new rules-in-form will be rigorously and consistently applied from the top down, eventually undermining all INGO activities in China. This scenario is made more likely by the perceived escalation of a "Cold War" between the US and China, which might continue under the Biden administration. Also, the recent Covid-19 pandemic has fostered security-oriented foreign policies and trends of national entrenchment in China as elsewhere, thus significantly restraining INGO activities in the short and maybe mid-term. The longer-term impacts of these factors, though, are still difficult to gauge.
The "recalibrating" scenario has proven plausible, too, to the extent that security concerns continue to compete with the social policy concerns of the party-state. As the empirical analysis has shown, there is evidence of at least partial changes in the rules-inuse and a gradual crowding out of INGOs by Chinese counterparts in the charity and social service sector. Unlike the "restraining" scenario, which assumes a consistent central policy agenda, the "recalibrating" scenario emphasises the continued ambiguities and competing agendas within the Chinese party-state. Its occurrence appears more likely in the mid and longer term under conditions of continued or revitalised mutual engagement and strong economic and political inter-dependencies.
Finally, the "reorienting" scenario assumes a continued dominant role for global norm-shaping ambitions within the CCP leadership, leading to a "new game in town" for the international non-profit sector. This scenario is empirically plausible to the extent that a diplomatic approach to INGOs prevails over the security apparatus' generalised distrust of foreign non-governmental actors. The analysis has shown that INGOs, while gradually losing their perceived usefulness as supporters of China's domestic development in the eyes of Chinese authorities, are increasingly meant to serve as international agents for "people-to-people connectivity" under the BRI. While we have presented evidence that spaces for INGO activities in and with China are indeed increasingly affected by the party-state's international "soft power" ambitions, the longer-term likelihood of the "reorienting" scenario rests on a high degree of interdependence and continued engagement between China and the Western countries that remain INGOs' main countries of origin. Such a new game for global civil society will also require the willingness of at least a significant share of INGOs to adjust their strategic orientation and play along under changing rules. While some INGOs are reluctantly doing so, continued research into the modalities of INGO co-operation with Chinese party-state actors and GONGOs will be necessary to assess the likelihood of these scenarios in the future.

1.
There is debate about the use and exact delineation of concepts such as NGOs, civil society organisations and third sector organisations, especially in the international context where binding legal definitions for non-profit organisations are absent. This article adopts INGO when referring to organisations that fall within the scope of the Overseas NGO Management