1 E.g. a new registration law in Russia, also designed to clamp down on the development of an independent social sphere, has created administrative hurdles for NGOs similar to those in China. Moscow Times, 24 August 2007. India has instituted a strict regulation system, evidence of a government wary of NGOs. Jalali, Rita, “International funding of NGOs in India,” Voluntas, Vol. 19 (2008), p. 172CrossRefGoogle Scholar.

2 Some argue that registration is crucial in helping Chinese environmental NGOs secure financial support and build transnational linkages. Ru, J. and Ortolano, L., “Development of citizen-organized environmental NGOs in China,” Voluntas, Vol. 20 (2009), pp. 141–68CrossRefGoogle Scholar.

3 Previous studies suggest that other social actors, such as entrepreneurs, prefer to formalize their relationship with the state to avoid negative government interference and increase economic opportunities. Dickson, Bruce, Red Capitalists in China: The Chinese Communist Party, Private Entrepreneurs, and Political Change (New York: Cambridge University Press, 2003)Google Scholar, and Wealth into Power: The Communist Party's Embrace of China's Private Sector (New York: Cambridge University Press, 2008)Google Scholar; Tsai, Kellee, Capitalism without Democracy: The Private Sector in Contemporary China (Ithaca, NY: Cornell University Press, 2007)Google Scholar.

4 Indicative of this, significant scholarly attention has been paid to environmental NGOs. E.g. Ho, Peter, “Greening without conflict? Environmentalism, NGOs and civil society in China,” Development and Change, Vol. 32 (2001)CrossRefGoogle Scholar; Lu, Y., “Environmental civil society and governance in China,” International Journal of Environmental Studies, Vol. 64, No. 1 (2007) pp. 59–69CrossRefGoogle Scholar; Ho, Peter and Edmonds, Richard L., China's Embedded Activism: Opportunities and Constrains of a Social Movement (New York: Routledge, 2008)Google Scholar; Mertha, Andrew, China's Water Warriors: Civic Action and Policy Change (Ithaca, NY: Cornell University Press, 2008)Google Scholar. Previous studies usually focused on a few organizations or offered in-depth case studies of individual groups. Attention to environmental NGOs has far outstripped studies of Chinese groups in the other issue areas (where work has only recently begun). There has been limited systematic comparison of groups across issue area.

5 Once research commenced, it became clear that most gay/lesbian NGOs addressed HIV/AIDS issues, suggesting overlap between two issue areas. Although this could introduce some bias (e.g. independence), it also demonstrates that issues acceptable to the state are limited; if an organization wants to enjoy political opportunities – and funding – it must significantly adapt.

6 I am careful to not generalize specific findings too far beyond these areas. However, insights drawn here might allow us to speculate (as I do in the conclusion) about registration as an example of complex state–society relations.

7 An appendix including a list of interviewees and the survey is available online at www.timothyhildebrandt.net/data.

8 To explain both registration and non-registration, this research could not rely on government data to establish a population (which would only include registered groups) and thus did not use a probability sample for the survey. The three issue areas were represented equally among responses. Interviewees and survey respondents were gathered through a snowball sampling method. I became aware of groups sometimes through independent directories. But because they are frequently (and often severely) outdated, I relied on personal interactions with leaders and observers, as well as listservs.

9 Howell, “Prospects for NGOs,” p. 8; Ho, “Greening without conflict?” p. 915.

10 Available online: http://huitong.mca.gov.cn/article/zcwj/200812/20081200023777.shtml.

11 This goal is difficult to achieve given that many sponsoring agencies lack the resources to supervise social organizations properly. J. Ru and L. Ortolano, “State control and environmental NGOs in China,” Association of Research on Nonprofit Organizations and Voluntary Action, Los Angeles, November 2004.

12 Ho, “Greening without conflict?” p. 902; Ma, Qiushi, “The governance of NGOs in China since 1978: how much autonomy?” Nonprofit and Voluntary Sector Quarterly, Vol. 31, No. 3 (2002), p. 306CrossRefGoogle Scholar.

13 Available online: http://cws.mca.gov.cn/accessory/200902/1233554233793.htm. Although the number of registered groups appears to have increased, figures are not reflective of an increase in the number of registered community-based organizations (minjian zuzhi); estimates include Party-affiliated mass organizations, industrial professional organizations and GONGOs.

14 Unregistered organizations are excluded from government estimates. However, based upon survey respondents, the number of social organizations in China, inclusive of unregistered groups, is likely to be larger than official figures.

15 At the very least, registration should not make an NGO more susceptible to government interference. But registration is not a panacea. It offers no guarantee that an organization will not encounter a strong negative state response. Others have pointed to similar problems, e.g. Lu, “Environmental civil society.”

16 Interview (IN) 23. Others disagree, noting that once a group becomes registered it is no longer a “grassroots organization” but instead more institutionalized and “governmental” (IN74, 75).

17 Others have found similarly high registration rates among environmental groups. Tang, Shuiyan and Zhan, Xueyong, “Civic environmental NGOs, civil society, and democratisation in China,” Journal of Development Studies, Vol. 44, No. 3 (2008), pp. 425–48CrossRefGoogle Scholar.

18 Significance refers to Pearson's chi-squared tests at the 90% confidence interval or higher.

19 Registration can be frustrating. An environmental NGO in Yunnan was denied registration after over a year of attempts because its name was deemed “too broad.” Tony Saich notes that regulations restrict organizations from using overly-broad names, which is the Ministry of Civil Affairs' (MoCA) attempt to preserve the monopoly of state-run social organizations (“Negotiating the state: the development of social organizations in China,” The China Quarterly, No. 161 (2000), pp. 123–41Google Scholar). Rather than change the name, the leader decided to forego registration. He reasoned that since he previously operated unregistered, he could continue to do so (IN31).

20 Other explanations included: “We already have good relations with the government”; “There is no pressure for us to be registered”; “It would cause us more problems, not fewer”; “The benefits do not outweigh the costs of the effort”; “Our work is more important than becoming registered.”

21 Although the survey did not find significant variation in difficulty of registration by geographic area, responses from the primary research sites are illuminating: among Yunnan-based organizations, 30% characterized registration as “about the same” as other areas and another 30% saw the process as “somewhat harder or harder” than other provinces. These responses support interview data which suggest Yunnan might not be as hospitable to NGOs as previously thought. Nearly 60% of Beijing-based respondents characterized the difficulty of legal registration in Beijing as “about the same” as in other provinces. Interviewees in Beijing usually assumed that central government dictates, as they pertained to NGOs in particular, were followed in other areas of the country; few foresaw a situation where local governments might be more strict than the central government in Beijing. It is unclear if leaders in one province are actually well-informed of the situation in others. This matters little, however, for showing that many leaders believe that government treatment of registration varies across the country.

22 Other determinants included: “good relations with government,” “adequate financial resources,” “co-operation with international NGOs,” “co-operation with domestic NGOs,” “a strong voice,” “contacts with media,” and “adequate capacity.”

23 HIV/AIDS NGOs were in between, with nearly a third of respondents ranking registration in the top three determinants of success, more than gay and lesbian groups, but less than environmental organizations.

24 It is likely that leaders' cavalier attitude towards registration is due to a lack of government attention (negative or otherwise). Were these groups experiencing more negative interactions with the state, they might believe that registration is a crucial mechanism to help them do the work they deem important.

25 Mary Gallagher has shown that autonomy from the state – which leaders associate with registration – decreases the influence of social actors. “China: the limits of civil society in a late Leninist state,” in Alagappa, M. (ed.), Civil Society and Political Change in Asia (Stanford, CA: Stanford University Press, 2004)Google Scholar.

26 IN15. Considerably fewer leaders resist registration because they believe it will actually tie them closer to the government (IN63).

27 IN29

28 IN26

29 She clarifies that the tactic is not for everyone: “If you have a good relationship with the government this is not a problem. But if you have a bad relationship it will not work” (IN64).

30 IN14, 35, 39. Many leaders from HIV/AIDS organizations report that local governments, with whom they often have pre-existing relationships, are unaware of registration requirements (IN24, 29).

31 IN35. While the 1998 regulations outline procedures for an organization to become registered, they does not indicate that failure to do so would elicit any particular punishment. However, ambiguities in the law do not mean it is open season for unregistered groups. Governments could use pre-existing laws and regulations (e.g. regarding publishing or state secrets) to exact punishment if they wish. Interestingly, there appears to be no qualitative difference in severity of punishment for groups that are registered and those that are not. The most common punishment for unregistered groups is the loss of political opportunity. When registered groups violate the regulation, however, their activities can be deemed illegal and they can face various punishments (e.g. leaders placed under travel bans, organizations stripped of registration status). Xinhua reported that the China Sexology Association was ordered to cease operations for six months because of violations to their registration status as non-profit-making. MoCA officials charged the organization with profiting from the sale of bronze “sponsorship plates” to manufacturers of sex health products (11 February 2008).

32 However, unregistered groups are significantly less likely to have formal meetings with central government officials: 67% reported meeting formally with these officials “rarely” or “never,” providing further evidence that unregistered groups enjoy closer relations with local officials than central government officials, who are more likely to prefer that all NGOs are registered.

33 58% of environmental groups are four years old or older, while only 27% and 29% of HIV/AIDS and gay and lesbian groups are in the same age category, respectively.

34 Predicted probabilities for HIV/AIDS and gay groups are indistinguishable, appearing as one line in Figure 3.

35 In 2004, a new regulation on foundations (defined as not-for-profit NGOs that rely on foreign and domestic donations for operation) relaxed this particular restriction. Leaders initially hoped to take advantage of this less restrictive regulation. However, NGOs have had a difficult time registering under foundation regulations; most are still subject to the 1998 social organization regulations.

36 Saich, “Negotiating the state.”

37 The central government is reportedly devising a plan to create a new bureaucracy tasked with co-ordinating, registering and controlling NGOs. An informant in Yunnan reported that local officials have used this report to put all registrations in the province on hold until it receives “policy clarifications” from Beijing (IN28). Therefore, even when the centre attempts to create more order, the local governments can use it to push their own interests, in this case, keeping groups unregistered.

38 IN14, 20

39 I am grateful to an anonymous reviewer for raising this point.

40 Schwartz, Jonathan and Shieh, Shawn, State and Society Responses to Social Welfare Needs: Serving the People (New York: Routledge, 2009)Google Scholar; Wang, S. and Sun, B., “Zhongguo minjian zuzhi fazhan gaikuang” (“Introduction to the development of civil organizations”), in Yu, K. et al. (eds.), Zhongguo gongmin shehui de xingqi yu zhili de bianqian (The Emergence of Civil Society and its Significance to Governance in Reform China) (Beijing: Social Science Documents Press, 2002), pp. 234–70Google Scholar.

41 IN27, 28

42 IN24, 25, 28, 63. No interviewee believed the government was reluctant to work with other groups because they represent individuals who engage in illegal behaviour. They insist it is simply an issue of economic opportunity.

43 One story widely circulated within the NGO community claims that a local CDC official in Heilongjiang, understanding that he needed a “civil society partner” to secure HIV/AIDS funds, enlisted his niece to start a gay men's group (IN38).

44 This offers another example of competition for limited resources common within the Chinese government. See Oi, Jean, “Fiscal Reform and the Economic Foundations of Local State Corporatism in China,” World Politics Vol. 45 (1992) pp. 99–124CrossRefGoogle Scholar.

45 IN66. The government sometimes matches international funders with agencies. In Yunnan, the US Agency for International Development, through two separate contractors, funded two different NGOs. In order to secure government sponsorship, one group linked up with a provincial level agency and the other with one at the city level. This not only led to competition and overlap between the groups, but also competition between the government agencies themselves (IN9).

46 The filter scheme requires a high number of government agencies at various be involved in dispersion. When recipients are unregistered, another set of hands through which funds must pass is added alongside even more opportunities for corruption.

47 IN43

48 IN13

49 IN50

50 IN13

51 IN53

52 IN52

53 Keck, Margaret and Sikkink, Kathryn, Activists beyond Borders (Ithaca, NY: Cornell University Press, 1998)Google Scholar.

54 Concerns about the lack of independence among recipients and corruption among lower-level government officials reached a head in 2010 when the Global Fund suspended the monies bound for China in May 2011. The suspension was lifted four months later when the Chinese government agreed to make changes. However, no details of changes have been released to the public. “Global Fund lifts China grant freeze,” Associated Press, 23 August 2011.

55 There is a meme among domestic NGOs that INGOs – particularly those where the government plays an important role – are too concerned about the government relations at their expense. An HIV/AIDS leader from Henan complained that on a recent visit by UNAIDS chief Peter Piet no NGOs were invited to participate (IN54). Although she partly blames the provincial government for the oversight, she believes the international organizations could go further in insisting that NGOs play a more prominent role in the province's battle against HIV/AIDS.

56 IN25

57 IN9

58 IN20

59 IN9, 24, 33

60 IN27

61 IN24

62 IN28

63 Given that registration is administered by MoCA at the central level, it is unlikely that the local government alone has the power to strip an NGO of its registration.