1113801 NVSXXX10.1177/08997640221113801Nonprofit and Voluntary Sector QuarterlyWang research-article2022 Articles What Matters for Charitable Donations Under Authoritarianism? An Examination of Organizational Legitimacy and Political Connections Nonprofit and Voluntary Sector Quarterly 1–30 © The Author(s) 2022 Article reuse guidelines: sagepub.com/journals-permissions https://doi.org/10.1177/08997640221113801 DOI: 10.1177/08997640221113801 journals.sagepub.com/home/nvs Qun Wang1 Abstract Voluntary certification that signals nonprofit organizations’ legitimacy has been found to be positively related to charitable donations in Europe and North America. However, whether these proxies of organizational legitimacy matter under authoritarian regimes where political connections are deemed critical for resource acquisition has not been explored. Based on a sample of 2,021 Chinese foundations, this article finds that organizational legitimacy achieved through certification is tied to an increase in charitable donations. Meanwhile, organizational legitimacy tends to negatively moderate the relationship between foundations’ political affiliation and charitable donations. That is, certification lifts the amount of charitable donations to civic foundations that are not formally affiliated with the state more than it does for politically affiliated foundations. As the first quantitative study on the effect of voluntary nonprofit certification programs on charitable donations outside Western democracies, this article contributes to our understanding of fundraising in China and other authoritarian states. Keywords organizational legitimacy, accreditation and certification, political connections, charitable donations, authoritarianism 1 The University of Toledo, OH, USA Corresponding Author: Qun Wang, Department of Political Science and Public Administration, The University of Toledo, Snyder Memorial Building 3007, MS 140, Toledo, OH 43606-3390, USA. Email: qun.wang@utoledo.edu 2 Nonprofit and Voluntary Sector Quarterly 00(0) Introduction Throughout the 1990s, articles about nonprofit wrongdoings began to permeate the U.S. press. Gibelman and Gelman (2004) identified theft, mismanagement of resources, excessive compensation, and so forth among U.S. nonprofits from media reports during the period 2001–2004. “The nonprofit sector’s claims to exist for the public good are no longer being taken on faith” (Brody, 2002, p. 472). Now donors demand verification, because it is difficult to know whether a charity’s request for support is legitimate (Princeton Survey Research Associates, 2001). The United States is not an isolated case. The press and academia have highlighted scandals of the nonprofit sector in various countries and requested proof for legitimacy. In response, pursuance of accreditation and certification for organizational legitimacy among nonprofits, initiated by either watchdog or government agencies, has been on the rise globally (Gugerty, 2009). Accreditation and certification are processes of assessment by an external organization that investigates the operation of an applicant (N. C. Feng et al., 2019; Ortmann & Myslivecek, 2010). They typically involve a review of organizations’ conformity, competence, and accountability against established standards. According to the signaling theory, nonprofits may utilize accreditation and certification to demonstrate their legitimacy (N. C. Feng et al., 2019). This way they differentiate themselves and convince stakeholders they deserve resources. Empirical research on nonprofits based in Europe and North America has supported a positive relationship between organizational legitimacy attested through accreditation and certification and donation income (e.g., Bekkers, 2003; N. C. Feng et al., 2019). Nonprofits have grown rapidly outside Western democracies in the last few decades. Realizing that a strong group of socially oriented organizations fundamentally align with regime interests, authoritarian states began to tolerate nongovernmental organizations (NGOs) independent of the state and incorporated capacity-building into their nonprofit policy. Meanwhile, in countries such as Russia, Turkey, and China, charitable donations have become a major source of nonprofit revenue.1 However, little research has focused on organizational legitimacy symbolized through sector-wide nonprofit accreditation and certification (for an exception, see Cheng & Wu, 2021). Under authoritarian rule, political institutions function in the state-controlled resource ecology. States are both governor and competitor in the nonprofit sector (Kang, 2018). Because charitable donation is limited, governments must be selective in their help and direct this source of revenue to key NGOs that serve as state agent and social actor at the same time. Not surprisingly, the literature usually acknowledges the environmental restrictions and investigates the impact of NGOs’ political connections on nonprofit revenue (e.g., Johnson & Ni, 2015; Ni & Zhan, 2017; Wei, 2020; Zhao et al., 2016). The rationale is that NGOs need to maintain strong and dependent relationships with the state for resources needed to advance their work. The findings of political connections’ tie to nonprofit revenue, however, became increasingly mixed. Wei (2020) suggested that in China NGOs’ political connections play an indirect role. Therefore, political connections may affect fundraising Wang 3 contingently through their interactions with other organizational attributes. Using 2,021 foundations in China—which are 66.72% of the entire population of foundations registered before 2013—this study helps fill the gaps by examining how organizational legitimacy (i.e., certification), political connections, and their interactions are related to charitable donations. China was chosen as the empirical setting for several reasons. In the current century, the Chinese party-state has gradually reduced its long-term funding commitment to NGOs, leaving many NGOs to rely on other sources of revenue. In comparison to other authoritarian states, China has more available data and empirical studies. Although China has its own peculiarities, distinct authoritarian regimes have exhibited surprisingly similar patterns in their nonprofit policy (Richter & Hatch, 2013). As the largest authoritarian state in the world, what is happening to Chinese NGOs is relevant for nonprofit sectors outside Western democracies. Findings from this study can directly inform prior literature and make three specific contributions. First, it introduces voluntary accreditation and certification to the study of nonprofit fundraising under authoritarianism. Prior studies used internal organizational variables not readily visible to potential donors, such as expertise, as legitimacy. Recently, Cheng and Wu (2021) found that the level of transparency among Chinese foundations, measured by the Foundation Transparency Index (FTI), was positively tied to donation income. Different from Cheng and Wu, this current study examines three certification programs that foundations voluntarily participate in rather than being passively rated. It also differs in that Cheng and Wu emphasized accountability through information disclosure, whereas this study focuses on regulatory conformity and the adoption of best practices. Second, by combining three data sources, this study creates a unique data set that measures political connections through organizational affiliation with the state in addition to board interlocking. Prior research usually operationalized political affiliation into a dichotomous variable so that an NGO was either politically affiliated or not (e.g., Li et al., 2017). Jia (2011) argued that the differentiation of NGOs’ political affiliation is hidden if we see Chinese NGOs this way. This study considers the nature, formation, and degree of NGOs’ affiliation and categorizes it according to governmental and quasi-governmental affiliation. How the two forms of political affiliation are tied to charitable donations may reveal nuances largely ignored in prior literature. This novel operationalization of political affiliation can be a useful methodological tool for future research. Finally, this study connects the organizational legitimacy literature advanced in Western societies and political determinants as propositions for authoritarian regimes by examining the moderating effect of organizational legitimacy on the relationship between political connections and charitable donations. Prior studies usually excluded the potential moderating effect of organizational attributes on political connections (e.g., Johnson & Ni, 2015) or considered the value of political connections as contingent on external shock (Ni & Zhan, 2017). In a rare exception, Cheng and Wu (2021) found that in China the presence of government officials on boards of directors helped foundations obtain more charitable donations only when foundations achieved roughly 4 Nonprofit and Voluntary Sector Quarterly 00(0) above 60% of the FTI scores. Alternatively, this study investigates whether the outcome of voluntary certification programs leverage the effect of both the number of government officials and the political affiliation on charitable donations. To a certain extent, this study contributes to the reconciliation of prior literature and a comprehensive understanding of government-NGO relations and fundraising under authoritarian rule. The rest of the article is structured as follows. First, it reviews the literature and proposes hypotheses. Next, it describes the research context, data, variables, and method. It then presents the results and discusses the implications. It concludes with a summary and limitations of the study. Literature Review Organizational Legitimacy Through Accreditation and Certification Organizational legitimacy is a generalized perception that the actions of an entity are desirable, proper, or appropriate within a system of norms, values, and beliefs (Suchman, 1995). Stakeholders perceive legitimate organizations as more trustworthy. Therefore, nonprofits may adopt endorsed practices into their formal structure (Powell & DiMaggio, 1991). Agency theory posits that to reduce risk, donors look for “insurance” that nonprofits allocate donated funds appropriately (Slatten et al., 2011). In a robust civil society, resource providers tend to provide resources to an NGO that possesses perceived legitimacy (Galaskiewicz & Bielefeld, 1998). NGOs are an emerging organizational field in most authoritarian societies. The public does not understand NGO operations very well (Henderson, 2011; Hsu & Hasmath, 2017). Moreover, donors encounter information asymmetry in the donation process and lack the expertise to thoroughly evaluate each NGO (Parsons, 2007). This is especially true where transparency is low in the nonprofit sector. In fact, low transparency has inhibited philanthropy in China and other transitioning countries (Wiepking & Handy, 2015). Table 1 presents three indexes rating nonprofit transparency in China. Each index similarly inspected the disclosure of organizational, governance, financial, and program information and gave low scores. A survey by the China Charity Information Centre reported that in 2013 about 80% of the people were unsatisfied with nonprofit transparency.2 Even in 2020, only 30% of the rated foundations disclosed financial reports, and 14.6% had little information available to the public (see Sources in Table 1). Proxies of organizational legitimacy that are easier to perceive and access, such as accreditation and certification, thus give donors useful references other than performance measures drawn from often hard-to-obtain NGO annual reports. Unlike their passive role when rated by a watchdog agency, NGOs actively seek to be accredited or certified for a score, status, or qualification (N. C. Feng et al., 2016). Accreditation and certification are signaling mechanisms with which nonprofits can demonstrate their quality. They are also a communication tool that mitigates donors’ lack of knowledge about an NGO. In the competitive fundraising market, an unaccredited or uncertified 5 Wang Table 1. Transparency of Chinese NGOs (Full Marks: 100 Points). Year 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Rater China Philanthropic Transparency Index (PTI) Grassroots NGO Transparency Index (GTI) Foundation Transparency Index (FTI) Average score Sample size Average score Sample size Average score 39 33 45.1 43.11 44.1 99 1,000 500 1,000 1,000 27.23 27.87 32.44 1,000 1,300 1,738 China Charity Information Centre Tsinghua University Sample size 48.89 2,508 43.91 2,604 49.45 3,053 48.91 3,637 47.73 4,234 33.67 4,956 50.08 5,694 51.34 6,394 55.43 2,478 China Foundation Center Source. China Foundation Center (http://fti.foundationcenter.org.cn/PDFFile/FTI2018Conferencesy.pdf; http://fti.foundationcenter.org.cn/PDFFile/FTI2019Conference-sy.pdf; http://fti. foundationcenter.org.cn/PDFFile/FTI2020.pdf); One Foundation (http://www.onefoundation.cn/index. php?m=article&a=show&id=879); Peng (2014). Note. The GTI does not include foundations. China Foundation Center rated only foundations having a website for the year 2020. NGO = nongovernmental organization. nonprofit that engages in best practices may not effectively distinguish itself from others that do not perform at the same standard. Bekkers (2003) discovered that in the Netherlands, donors had higher trust in charities and gave more money to charitable causes if they knew the CBF-Keur accreditation system. Using a sample of 102 U.S. nonprofits, N. C. Feng et al. (2016) found that the Standards for Excellence certification was associated with increases in donations relative to a control group of nonprofits that did not receive the certification. Adena et al. (2019) implemented an experiment to test the impact of quality certification on donations to charities. The result showed that participants presented with a certified charity chose higher donations by 10% compared with the control group that was not given such information. Considering the relatively low nonprofit transparency in China and other authoritarian states, accreditation and certification may offer NGOs a similar strategic advantage for fundraising. Therefore, I submit the following hypothesis: Hypothesis 1: Organizational legitimacy attested through nonprofit accreditation or certification is positively related to the amount of charitable donations to NGOs. 6 Nonprofit and Voluntary Sector Quarterly 00(0) Political Connections Under authoritarian rule where the state usually retains an important role in deciding resource distribution for NGOs, political connections have been charged with bringing about financial resources (e.g., Henderson, 2011; Q. Ma, 2006). Prior literature commonly studied the value of political connections for nonprofit fundraising in China within the resource dependence framework. Cheng and Wu (2021) and Zhou and Ye (2021) posited that political connections would enhance NGOs’ legitimacy, which in turn could lead to more donation income. Yet scandals that occurred in politically connected NGOs frequently made headlines. Johnson and Ni (2015) and Wei (2020) formulated political connections as guanxi, a Chinese culture broadly defined as an informal interpersonal relationship. However, guanxi is hidden and hard to examine (Zhou and Ye, 2021) and does not necessarily capture the impact of formal political embeddedness (Q. Wang, 2021b). Ni and Zhan (2017) and Wei (2020) equated political connections with government control and believed such control meant greater financial opportunities for NGOs. Their argument was theoretically grounded but unclear about how government control can be translated into charitable donation—a source of nonprofit revenue not in the possession of the party-state. To understand the effect of political connections on fundraising, this study adopts an alternative approach in order to institutionally and practically examine the mechanism through which the state channels charitable donations to NGOs. Zhang et al. (2016) observed that a large amount of corporate donations went to politically connected NGOs. They attributed this phenomenon to China’s weak market institutions. On one hand, the Chinese government co-opts business elites by two means. First, the state maintains control of key resources and uses them to incentivize conformity. Second, the state seeks to incorporate business elites into the political system by appointing them as delegates to legislative bodies. On the other hand, the state extends its grabbing hand into the pockets of firms for charitable donations. Government officials may solicit donations through both persuasion and intimidation (L. Feng & Zhang, 2014). The government is more likely to give preferential treatment to firms that make such contributions than to those that do not (H. Wang & Qian, 2011). At all administrative levels, the Chinese government frequently forces its state-salaried employees to donate to government-run charitable projects and politically connected NGOs. The donation can be a flat rate or in accordance with the employee’s administrative rank. In 2005, many cities began launching the “donate one yuan a day” campaign among state-salaried employees. Some local governments even deducted the donation directly from employees’ paychecks. Those cases are prevalent rather than sporadic (L. Feng & Zhang, 2014).3 Drawing data from the 2012 Chinese General Social Survey, Hu and Kong (2021) found that in comparison with people working in the private sector, employees salaried by the state and state-owned enterprises were more likely to donate when the donation was either suggested or coerced by government and less likely to donate when the donation was voluntary. In her analysis of government-NGO relations, Xu (2008) identified three ways in which the government helped the government-organized China Population Welfare Wang 7 Foundation (CPWF) fundraise. First, government officials sometimes personally availed themselves of fundraising activities. Second, the CPWF and its supervisory agency—the State Population Office—mobilized government and public service unit (shiye danwei) employees to donate. Third, the CPWF activated the State Population Office’s more than one million branches and 80 million individual and organizational members across the country to fundraise. Similarly, three fourths of the surveyed Russian philanthropists reported experiencing pressure from local authorities to donate to designated recipients (Livshin & Weitz, 2006). In Vietnam, compulsory giving includes (1) government employees are required to contribute to government-affiliated charities, (2) households and private businesses are visited by local officials and mass organizations for charitable contributions, and (3) community members are asked to support local charities on an annual or biannual basis (Nguyen & Doan, 2015). In these cases, political connections grant NGOs the privilege of raising donations that they otherwise would not have raised. Therefore, I submit the following hypothesis: Hypothesis 2: Political connections are positively related to the amount of charitable donations to NGOs. Moderating Effect of Organizational Legitimacy Against the backdrop of interrupted associational life and a stagnant economy, in the late 1970s, Chinese NGOs emerged as the state’s policy coordination intermediaries and fundraising arms in a top-down manner. Three factors rendered competition for nonprofit resources almost irrelevant: the small number of NGOs, the large employment in public organizations and state-owned enterprises, and prevention of a second NGO working on similar issues from registering in each administrative jurisdiction. Organizational legitimacy, if any, would be the leadership of “old revolutionaries” and their spouses or children who enjoyed the admiration of the people who retained a fresh memory of the Anti-Japanese War (Q. Wang, 2021a). Such leadership, however, did not equate conformity, competence, or accountability. With the introduction of the concept of “NGO” to the Chinese society in the late 1990s, more and more people and groups began seeking NGO registration. Among the founders, some had been government officials who used their institutional experience to build partnerships with the state (Hsu & Jiang, 2015). Other founders, especially those without such experience, often pursued political embeddedness by instituting themselves in a stateendorsed organization or by inviting government officials to their boards (Ho, 2007). These political connections were believed to place NGOs in an advantageous position in various ways. Chinese NGOs grew rapidly in the 2000s. Concurrently, the burgeoning private economy relaxed public sector employment and people’s dependence on the state. Political leadership in the nonprofit sector no longer enjoyed the legitimacy derived from the war times. As nonprofit sector scandals made headlines, China began enacting regulations to standardize the nonprofit sector like other authoritarian states 8 Nonprofit and Voluntary Sector Quarterly 00(0) (Henderson, 2011; Richter & Hatch, 2013). NGOs needed to adapt to the new environment in which they competed for limited nonprofit resources by, for example, strengthening legitimacy other than, if not on top of, political connections.4 Newly emerged accreditation and certification programs that signal legitimacy can potentially moderate the effects of political connections. While all NGOs can voluntarily pursue accreditation and certification, which in turn increases their competitiveness in fundraising, the effect could be stronger for NGOs that have no or weak political connections. That is, being accredited or certified would reduce the gap between the amounts of charitable donations attributable to political connections. First, politically connected NGOs are more likely to mimic governmental structure and operate in a centralized, bureaucratic manner (Toepler & Fröhlich, 2020) that alienates potential donors. For example, Nubo Huang, a Chinese entrepreneur said he did not want to donate to government-organized NGOs for the above reasons.5 Moreover, in the digital age, the internet sometimes fosters an anti-authority culture. NGOs showing off their political connections “may be inviting more enemies than donors” (Zhou & Ye, 2021, p. 501). Second, when having no or weak political connections, NGOs could operate free from direct government control and have more autonomy in planning their boards, programs, and fundraising. They are more likely to network with other organizations to gain information and broker resources (Li et al., 2018). Without a commandable donor base, they have no option but to maximize the value of accreditation or certification to motivate potential donors, especially small donors who have become increasingly important and who pay attention to organizational legitimacy (Zhou & Ye, 2021). As a result, although accreditation and certification enhance NGOs’ legitimacy sectorwise, they benefit NGOs differently. Put another way, organizational legitimacy attested through accreditation and certification may discount the value of political connections in nonprofit resource acquisition, offering NGOs with no or weak political connections an alternative funding opportunity. Therefore, I submit the following hypothesis: Hypothesis 3: Organizational legitimacy attested through nonprofit accreditation or certification negatively moderates the relationship between political connections and the amount of charitable donations to NGOs. Method Research Context Chinese foundations are the subjects under study. By the end of 2019, 15 years after the promulgation of the 2004 Regulations for the Management of Foundations, the number of foundations had reached 7,585 (see Table 2). Charitable donations to Chinese foundations increased from 34.05 billion yuan to 67.71 billion yuan from 2010 to 2019. They surpassed the China Charity Federation system to receive the most charitable donations from no later than 2010. During 2008–2013, donations accounted for approximately 84% of their total annual revenue (Q. Wang & He, 2018). Albeit only one segment of China’s nonprofit sector, foundations’ similarities with other forms of NGOs are more salient than their differences. 9 Wang Table 2. Charitable Donations Received by Chinese Foundations. Year Number of foundations Donation to foundations (in billion yuan) Donation to foundations / National total 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 892 975 1,144 1,340 1,597 1,843 2,202 2,614 3,029 3,549 4,117 4,784 5,559 6,307 7,034 7,585 34.05 33.70 30.57 37.35 38.32 44.56 62.55 65.80 64.59 67.71 33.00% 39.88% 37.42% 37.74% 36.76% 40.19% 44.91% 43.87% 44.88% 44.86% Source. China Charity Federation (http://www.cncf.org.cn/storage/ueditor/file/2020/09/19/39c9155e41612 e431eb076253d1777ec.pdf); China Statistical Yearbooks; Q. Wang (2020). Note. The exchange rate between U.S. dollars and Chinese yuan during this period was roughly 1:6.5. Data and Sample The study used data from three data sources, resulting in a sample of 2,021 foundations or 66.72% of the foundations established before 2013.6 The first source was the Research Infrastructure of Chinese Foundations (RICF), which holds records of 3,342 foundations’ organizational, personnel, financial, and program information for the year 2013 (see the data descriptor by J. Ma et al., 2017). The other two sources were Q. Wang’s (2018a) data for 4,229 foundations’ link with the state and Q. Wang’s (2018b) typological study of 3,342 foundations, in which the foundations were divided into government-created and other groups. Foundations registered in 2013 were removed from the sample because some foundations registered that year reported donation income for the purpose of registration. These numbers did not reflect a foundation’s fundraising ability under normal operation. Dependent Variable The dependent variable was the amount of charitable donations a foundation received in 2013. Because the distribution of charitable donations was skewed, I took the natural logarithm of this variable (see the histogram in the Supplemental Online Material). 10 Nonprofit and Voluntary Sector Quarterly 00(0) Independent Variables Organizational legitimacy. The Chinese government administratively established three voluntary certification programs for NGO legitimacy in the 2000s. The first was the nonprofit evaluation (shehui zuzhi pinggu). It is a comprehensive assessment of social organizations’ (what NGOs are called in China) basic organizational condition (jichu tiaojian), governance (neibu zhili), effectiveness (gongzuo jixiao), and societal recognition (shehui pingjia) measured by comments from internal and external stakeholders. Each administrative jurisdiction sets up an expert committee comprising government officials, researchers, lawyers, accountants, and practitioners to conduct on-site evaluation. Then a review committee further examines the evaluation result. The nonprofit evaluation rates NGOs from 1A to 5A. A higher score means better performance; 3A and above indicate the government’s recognition of an NGO’s performance. According to the RICF, by 2013 roughly 20% (646 out of 3,342) of the foundations had self-selected to participate in this evaluation, and about 17.5% (or 586) were awarded 3A or above. Although the participation rate was relatively low in 2013, a good evaluation score could still provide positive signals to potential donors. The results remain effective for 5 years. The other two certification programs concern nonprofit tax qualification. In China, NGOs do not automatically obtain nonprofit tax qualifications upon registration; they must apply. In the 2000s, the Chinese government began to regulate nonprofit tax exemption (charitable donations being nontaxable) and pretax deduction (deduction of donation from donors’ income before tax). Before applying for the tax exemption qualification, NGOs must have demonstrated conformity to financial requirements, such as proper use of funds and reasonable salary and benefits. The pretax deduction qualification highlights operational conformity. For example, an administrative penalty disqualifies an NGO for this qualification for 3 years. The two tax qualifications offer incentives to donors. The nonprofit tax exemption qualification indicates a possible higher proportion of donations for program implementation. The RICF reports that in 2013 about 56% (or 1,886) of the foundations had obtained this qualification. It also reports that in the same year about 21% (or 717) of foundations had obtained the nonprofit pretax deduction qualification. At that time, the Enterprise Income Tax Law permitted donations of up to 3% of firms’ profits to be deductible in the calculation of their taxable income. The two tax qualifications are effective for 5 years and 3 years, respectively. The certifying agencies must publicize nonprofit evaluation scores and tax qualifications on official websites. The three certification programs are visible proxy for organizational legitimacy in China’s nonprofit sector. Nonprofit evaluation score was coded as an interval variable ranging from 0 to 5, where 0 meant a foundation had not been rated yet and 1–5 corresponded to 1A–5A. This variable denotes the extent to which an NGO has proved its legitimacy via certification. The two tax qualifications each were coded as a dummy variable, where 1 meant a foundation had obtained the qualification and 0 if not. Wang 11 Political connections. Prior literature usually operationalizes political connections in two ways. First is the number or presence of government officials on an NGO’s board of directors (e.g., Cheng & Wu, 2021; Johnson & Ni, 2015). This study adopted the number of government officials (see a discussion on the presence of government officials in the Supplemental Online Material). Second is to code NGOs’ political connections as their formal affiliation with the state—either existent or nonexistent. Such operationalization does not fully address the fact that the nature, formation, and degree of political affiliation may result in different dynamics between the state and NGOs. Presently in China, both government agencies and public service units (PSUs) are frequent founders of NGOs. The term government agencies refers to administrative agencies, legislatures, law enforcement agencies, mass organizations, and democratic parties submitting to the leadership of the Chinese Communist Party (CCP) (Ni & Zhan, 2017; Q. Wang, 2018a). PSUs are quasi-government agencies under the purview of the state (Pearson, 2007). Typical PSUs include universities, hospitals, parks, and museums. They provide services in education, science and technology, culture, and sanitation using state-owned assets. The state controls PSUs in several ways. First, the state is the de facto sole creator of PSUs. The Interim Regulation on the Registration of Public Service Units (2004) allows “other organizations” to establish PSUs with state-owned assets, but it is practically impossible to find PSUs established by other organizations. Second, executives of these agencies are usually appointed by a corresponding CCP organization department. They bear comparable administrative ranks to their peers in government agencies and are transferable from and to governmental positions. Third, PSU employees are often salaried by the state and depend on the state for inexpensive housing, health care, and pensions. PSUs are part of the state and complement the government’s work. Kang (2018) made it explicit that political connections are related to the government’s treatment of NGOs, including resources, trust, and duties. He noted that the new blueprint for the intended relations between the party-state and businesses, nonprofits, and individuals is that the party-state is in the center and the rest are “ranked in order as a series of concentric rings. . . . [T]his hierarchical civil administration system was under the charge of the ruling party and was operated by the government” (p. 4). Figure 1 graphically illustrates such relations. Quasi-government agencies are common in many countries. For example, the Russian government has decentralized tasks to quasi-government agencies like those in China in that those agencies are under the auspices of the state and complement the government’s work (Peregudov & Semenenko, 1996). Since quasi-government and government agencies bear different functions and closeness to the state, distinguishing between NGOs’ affiliation with the state is more revealing than regarding them as homogeneous government-organized NGOs. Political affiliation was therefore coded into a categorical variable that took the value of 1 for foundations being civic, meaning the absence of formal affiliation with the state; 2 for quasi-governmental affiliation; and 3 for governmental affiliation. Civic foundations were treated as the reference category. 12 Nonprofit and Voluntary Sector Quarterly 00(0) Figure 1. Political connections and the distribution of resources, trust, and duties. Note. Grassroots NGOs in this figure refers to unregistered NGOs. Control Variables Organizational variables. Age represents the years a foundation has existed. Previous research shows that age could be positively tied to the amount of charitable donations a nonprofit received (Fischer et al., 2011). A larger board size may increase the board’s access to external resources (Guo, 2007). Professionalization is important for fundraising (Hwang & Powell, 2009) and is measured by the number of full-time employees. I took the natural logarithm of full-time employees to reduce skewness. The Regulations for the Management of Foundations classified foundations as public or nonpublic upon registration. Only the former, given their public fundraising qualification, can fundraise from the public. The attainment of this qualification was coded as 1; otherwise 0.7 The Regulations also stipulated a registration system that defines foundations’ scope of operation as either national (coded as 1) or subnational (coded as 0). If a foundation works on issue areas enumerated in the Charity Law such as poverty alleviation, education, health, and so on, it was coded as 1; otherwise 0. Finally, university foundations developed rapidly and attracted large amounts of donations from alumni and entrepreneurs. To account for their special status in the nonprofit resource market, university foundations entered the regression in the value of 1; otherwise 0.8 Financial variables. Saxton and Wang (2014) found that fundraising expenditure is positively related to nonprofits’ donation income. Total assets 2012 measured organizational Wang 13 size. Fischer et al. (2011) found a negative relationship between nonprofits’ total assets of the preceding fiscal year and charitable donations. The two financial variables were logged due to high skewness. Contextual variables. The first was the 2013 nonprofit policy environment index developed by the China Philanthropy Research Institute (Z. Wang et al., 2016), which rates the supportiveness of nonprofit policy in each province. Economic development may affect resource availability. The 2013 provincial GDP per capita captured regional economy. Analytical Strategy This study adopted the ordinary least squares (OLS) regression as Johnson and Ni (2015) and Ni and Zhan (2017) did to log-transformed charitable donations that Chinese foundations received. In addition, the double-hurdle model, among two other alternatives, was implemented to account for the fact that a nontrivial portion of the sampled foundations—20.3% or 410 in number—reported a zero amount of donation income. The Tobit model assumes that zero observations are standard corner solutions, requiring that the receipt and amount of charitable donations be determined by the same process. The Heckman model assumes foundations that were selected to be recipients would receive a nonzero amount. The double-hurdle model nests the above two censoring mechanisms together (Li et al., 2020) and relaxes their assumptions (Forbes & Zampelli, 2011). Its specification undertakes a two-step estimation procedure: probit regression in the first step and linear regression in the second step. It requires two hurdles be overcome to observe positive values. Stated in terms of fundraising, a foundation must cross the first hurdle (1H) to be identified as a potential recipient and then the second hurdle (2H) to receive nonzero. The model allows the same factor to affect the receipt and amount of charitable donations in different ways. This flexibility accords with the reality that a potential recipient identified in the first step may receive a zero donation in the second step. In the following analysis, both OLS and double-hurdle models used robust standard errors to correct for heteroscedasticity. Results Table 3 provides the descriptive statistics. Table 4 presents the regression results. The OLS model (Model 1) showed that the three organizational legitimacy variables were positively related to charitable donations at p < .05 or lower. In the doublehurdle model (Models 2–3), nonprofit evaluation score was not tied to the probability of receiving charitable donations but was positively tied to the amount that donationreceiving foundations received at p < .05. Nonprofit tax exemption qualification and pretax deduction qualification predicted greater probability and amount of charitable donations at p < .05 or lower. Political connections, namely the number of government officials and quasi-governmental and governmental affiliations, were insignificant in both models. To assess the regressors in the double-hurdle model, it is helpful 14 Nonprofit and Voluntary Sector Quarterly 00(0) Table 3. Descriptive Statistics. Variable Charitable donations Charitable donations (logged) (H1) Nonprofit evaluation score (H1) Nonprofit tax exemption (H1) Nonprofit pretax deduction (H2) Number of government officials (H2) Political affiliation 1-Civic 2-Quasi-governmental 3-Governmental Age Board size Full-time employees Full-time employees (logged) Public fundraising qualification Scope of operation Issue area University foundation Fundraising expenditure Fundraising expenditure (logged) Total assets 2012 Total assets 2012 (logged) Nonprofit policy environment GDP per capita Observation Mean 2,021 2,021 2,021 2,021 2,021 1,928 12,491 6.094 0.979 0.271 0.701 0.573 2,021 599 408 1,014 2,021 1,911 2,019 2,019 2,021 2,021 2,021 2,021 2,021 2,021 2,021 2,021 2,021 2,021 2.205 7.808 13.114 3.714 1.144 0.454 0.080 0.880 0.117 42.28 0.650 37,876 9.208 0.486 61.923 Standard deviation 70,203 3.590 1.654 1.473 Minimum Maximum 0 0 0 0 0 0 1 7.458 6.480 6.205 0.863 578.44 1.531 150,599 1.423 0.164 21.395 1 5 0 0 0 0 0 0 0 0 3.69 1.545 0.110 22.921 1,608,360 14.291 5 1 1 16 3 32 25 111 4.718 1 1 1 1 18,962 9.850 2,871,878 14.87 0.803 99.607 Note. All monetary values are in 1,000 Chinese yuan. to analyze their marginal effects (see the last column in Table 4), which can be compared with OLS coefficients that are directly interpreted as marginal effects (Newman et al., 2003). In terms of the impact on the dependent variable, the organizational legitimacy and political connections variables exhibited similar significance and weight between the two regression methods. Therefore, Hypothesis 1 was supported; Hypothesis 2 was not. Tables 5–7 present the interaction effects between organizational legitimacy and political affiliation. All three OLS models (Models 4, 7, and 10) showed negative, significant interaction terms between each organizational legitimacy variable and the two forms of political affiliation at p < .1 or lower. According to Models 5 and 6, the interaction between nonprofit evaluation score and quasi-governmental affiliation reported a negative effect only on the probability of receiving charitable donations at p < .1. The interaction between nonprofit evaluation score and governmental affiliation reduced the 15 Wang Table 4. OLS and Double-Hurdle Models for Charitable Donations (Logged). Double-hurdle Organizational legitimacy (H1) Nonprofit evaluation score (H1) Nonprofit tax exemption (H1) Nonprofit pretax deduction Political connections (H2) Government officials (H2) Quasi-governmental affiliation (QA) (H2) Governmental affiliation (GA) Age Board size Full-time employees (logged) Public fundraising qualification Scope of operation Issue area University foundation Fundraising expenditure (logged) Total assets 2012 (logged) Nonprofit policy environment GDP per capita Constant R2 F Pseudo R2 Chi2 OLS 1H: Receipt 2H: Amount Marginal effects Model 1 Model 2 Model 3 dy/dx 0.103* (0.052) 0.582*** (0.180) 1.052*** (0.178) 0.027 (0.026) 0.211* (0.099) 0.272*** (0.079) 0.060* (0.029) 0.293** (0.103) 0.636*** (0.106) 0.096† (0.051) 0.603*** (0.184) 1.007*** (0.169) −0.072 (0.051) 0.095 (0.263) −0.126 (0.291) −0.060*** (0.012) 0.031* (0.013) 0.330** (0.108) 0.474† (0.282) 1.239*** (0.311) 1.183*** (0.275) 1.577*** (0.280) 0.293*** (0.047) 0.279*** (0.079) 0.459 (0.494) 0.009* (0.004) −0.080 (0.725) 0.226 40.310*** −0.022 (0.024) 0.178 (0.142) −0.097 (0.132) −0.017** (0.005) 0.014* (0.006) 0.086† (0.051) 0.099 (0.126) 0.297† (0.176) 0.223* (0.113) 0.763*** (0.208) 0.133*** (0.034) −0.058† (0.033) −0.017 (0.254) 0.004* (0.002) 0.435 (0.321) −0.032 (0.028) −0.230 (0.165) 0.063 (0.175) −0.033*** (0.007) 0.006 (0.007) 0.264*** (0.059) 0.307† (0.172) 0.691*** (0.178) 0.933*** (0.172) 0.763*** (0.178) 0.120*** (0.027) 0.492*** (0.053) 0.355 (0.303) 0.002 (0.002) 0.858† (0.486) −0.065 (0.048) 0.087 (0.256) −0.124 (0.283) −0.057*** (0.011) 0.029* (0.012) 0.367*** 0.103 0.432 (0.265) 1.071*** (0.309) 1.140*** (0.248) 1.714*** (0.260) 0.332*** (0.062) 0.299*** (0.077) 0.259 (0.511) 0.010* (0.004) 0.106 907.280*** Note. Observation: 1,824. dy/dx for factor levels is the discrete change from the base level. OLS = ordinary least squares; QA = quasi-governmental affiliation; GA = governmental affiliation; GDP = gross domain product. † p < .1. * p < .05. ** p < .01. *** p < .001. 16 Nonprofit and Voluntary Sector Quarterly 00(0) probability at p < .01 but increased the amount to donation-receiving foundations at p < .05. In Models 8 and 9, the significant interaction term that led to lower probability of receiving charitable donations could be found only between nonprofit tax exemption qualification and governmental affiliation at p < .05. Models 11 and 12 reported two negative, significant interaction terms for the receipt of charitable donations between nonprofit tax deduction qualification and quasi-governmental and governmental affiliation at p < .01 or lower, respectively. In Models 10 and 11, the coefficients for quasigovernmental affiliation were positive, significant at p < .001. Interaction terms in these two models might indicate pretax deduction qualification’s offsetting effect on such affiliation. None of the interaction terms between the number of government officials and the three organizational legitimacy variables was reliably significant at p < .1 (see the Supplemental Online Material). Thus, Hypothesis 3 received limited support. Yet it is difficult to understand the interaction effects based on the coefficients alone since the interaction analysis involved multiple regression methods and both the probability and size of charitable donations. Therefore, it is beneficial to visualize these interaction terms (see Figures 2 to 4; also see predictive margins of political affiliation in the Supplemental Online Material) and interpret the marginal effects of political affiliation conditional on organizational legitimacy (see Table 8). As shown by the OLS and double-hurdle models in Figure 2, political affiliation tended to make a negative difference when nonprofit evaluation score was relatively high. The evidence is strong for the comparison between governmental and civic foundations that were awarded 4A or 5A status. Figure 3 shows that governmental foundations who had obtained the nonprofit tax exemption qualification received a significantly lower amount of charitable donations than civic foundations. Figure 4 reveals that without the nonprofit pretax deduction qualification, quasi-governmental foundations would receive more charitable donations than civic foundations. However, after these two types of foundations obtained this qualification, the gap became insignificant. In sum, nonprofit certification programs can lift the amount of charitable donations to civic foundations more than to politically affiliated foundations to the extent that (1) civic foundations would receive more charitable donations than governmental foundations or (2) the gap between civic and quasi-governmental foundations was greatly reduced. This pattern, however, did not reliably exist for the number of government officials (see marginal effects of the number of government officials in the Supplemental Online Material). The results for control variables were relatively reliable across all models. Full-time employees (logged), public fundraising qualification, scope of operation, issue area, university foundation, and fundraising expenditure (logged) all predicted more charitable donations at p < .1 or lower, consistent with the expectation in the OLS and doublehurdle models. Age was negatively tied to charitable donations. Future research can help determine whether younger foundations have implemented more innovative fundraising strategies. Total assets 2012 (logged) were positively related to the amount of charitable donations. Probably instead of penalizing those that had been accumulating wealth, Chinese donors believed larger foundations were more credible. However, this variable 17 Wang Table 5. Interactions Between Nonprofit Evaluation Score and Political Affiliation. Double-hurdle Organizational legitimacy (H1) Nonprofit evaluation score (H1) Nonprofit tax exemption (H1) Nonprofit pretax deduction Political connections (H2) Government officials (H2) Quasi-governmental affiliation (QA) (H2) Governmental affiliation (GA) Interactions (H3) Political Affiliation × Evaluation QA × Nonprofit Evaluation Score GA × Nonprofit Evaluation Score Age Board size Full-time employees (logged) Public fundraising qualification Scope of operation Issue area University foundation Fundraising expenditure (logged) Total assets 2012 (logged) Nonprofit policy environment GDP per capita Constant R2 F Pseudo R2 Chi2 OLS 1H: Receipt 2H: Amount Model 4 Model 5 Model 6 0.240*** (0.072) 0.580*** (0.180) 1.048*** (0.179) 0.16*** (0.050) 0.209* (0.100) 0.273*** (0.079) 0.002 (0.046) 0.286** (0.103) 0.626*** (0.106) −0.073 (0.051) 0.276 (0.280) 0.035 (0.311) −0.023 (0.024) 0.264† (0.149) 0.019 (0.137) −0.032 (0.028) −0.248 (0.172) −0.042 (0.185) −0.216* (0.099) −0.180† (0.098) −0.059*** (0.013) 0.031* (0.013) 0.333** (0.107) 0.480† (0.282) 1.252*** (0.312) 1.193*** (0.277) 1.600*** (0.281) 0.297*** (0.047) 0.278*** (0.079) 0.434 (0.493) 0.009* (0.004) −0.184 (0.728) 0.227 37.490*** −0.139† (0.084) −0.179** (0.057) −0.016** (0.005) 0.014* (0.006) 0.096† (0.051) 0.111 (0.127) 0.305† (0.179) 0.237* (0.113) 0.776*** (0.206) 0.139*** (0.034) −0.058† (0.033) −0.014 (0.255) 0.004† (0.002) 0.348 (0.320) 0.012 (0.065) 0.111* (0.056) −0.035*** (0.008) 0.006 (0.007) 0.262*** (0.059) 0.293† (0.172) 0.660*** (0.179) 0.922*** (0.171) 0.758*** (0.177) 0.119*** (0.026) 0.498*** (0.053) 0.342 (0.303) 0.003 (0.002) 0.871† (0.489) 0.107 921.860*** Note. Observation: 1,824. OLS = ordinary least squares; QA = quasi-governmental affiliation; GA = governmental affiliation; GDP = gross domain product. p < .1. * p < .05. ** p < .01. *** p < .001. † 18 Nonprofit and Voluntary Sector Quarterly 00(0) Table 6. Interactions Between Nonprofit Tax Exemption Qualification and Political Affiliation. Double-hurdle Organizational legitimacy (H1) Nonprofit evaluation score (H1) Nonprofit tax exemption (H1) Nonprofit pretax deduction Political connections (H2) Government officials (H2) Quasi-governmental affiliation (QA) (H2) Governmental affiliation (GA) Interactions (H3) Political Connections × Exemption QA × Nonprofit Tax Exemption GA × Nonprofit Tax Exemption Age Board size Full-time employees (logged) Public fundraising qualification Scope of operation Issue area University foundation Fundraising expenditure (logged) Total assets 2012 (logged) Nonprofit policy environment GDP per capita Constant R2 F Pseudo R2 Chi2 OLS 1H: Receipt 2H: Amount Model 7 Model 8 Model 9 0.102* (0.052) 1.044*** (0.27) 1.051*** (0.178) 0.027 (0.026) 0.466** (0.160) 0.271*** (0.079) 0.060* (0.029) 0.411* (0.161) 0.637*** (0.105) −0.075 (0.051) 0.291 (0.299) 0.076 (0.316) −0.025 (0.024) 0.241 (0.158) −0.014 (0.138) −0.032 (0.028) −0.170 (0.179) 0.121 (0.188) −0.614† (0.371) −0.774* (0.389) −0.058*** (0.013) 0.032* (0.013) 0.338** (0.108) 0.481† (0.284) 1.261*** (0.315) 1.217*** (0.276) 1.579*** (0.281) 0.289*** (0.047) 0.279*** (0.079) 0.471 (0.499) 0.009* (0.004) −0.284 (0.737) 0.227 36.620*** −0.248 (0.298) −0.443* (0.205) −0.016** (0.005) 0.014* (0.006) 0.091† (0.051) 0.105 (0.126) 0.313† (0.179) 0.238* (0.113) 0.759*** (0.209) 0.133*** (0.034) −0.057† (0.033) 0.002 (0.259) 0.004* (0.002) 0.349 (0.324) −0.171 (0.241) −0.193 (0.219) −0.033*** (0.007) 0.007 (0.007) 0.267*** (0.060) 0.308† (0.172) 0.693*** (0.181) 0.943*** (0.172) 0.763*** (0.178) 0.119*** (0.027) 0.492*** (0.053) 0.353 (0.306) 0.002 (0.002) 0.800 (0.498) 0.106 908.800*** Note. Observation: 1,824. OLS = ordinary least squares; QA = quasi-governmental affiliation; GA = governmental affiliation; GDP = gross domain product. † p < .1. * p < .05. ** p < .01. *** p < .001. 19 Wang Table 7. Interactions Between Nonprofit Pretax Deduction Qualification and Political Affiliation. Double-hurdle Organizational legitimacy (H1) Nonprofit evaluation score (H1) Nonprofit tax exemption (H1) Nonprofit pretax deduction Political connections (H2) Government officials (H2) Quasi-governmental affiliation (QA) (H2) Governmental affiliation (GA) Interactions (H3) Political Connections × Deduction QA × Nonprofit Pretax Deduction GA × Nonprofit Pretax Deduction Age Board size Full-time employees (logged) Public fundraising qualification Scope of operation Issue area University foundation Fundraising expenditure (logged) Total assets 2012 (logged) Nonprofit policy environment GDP per capita Constant R2 F Pseudo R2 Chi2 OLS 1H: Receipt 2H: Amount Model 10 Model 11 Model 12 0.096† (0.052) 0.562** (0.179) 1.851*** (0.327) 0.027 (0.026) 0.199* (0.099) 0.658*** (0.142) 0.058* (0.029) 0.292** (0.103) 0.650*** (0.188) −0.071 (0.051) 1.272*** (0.388) 0.487 (0.412) −0.022 (0.024) 0.770*** (0.219) 0.181 (0.167) −0.032 (0.028) −0.116 (0.242) 0.032 (0.243) −1.704*** (0.417) −0.891* (0.397) −0.058*** (0.012) 0.032* (0.013) 0.328** (0.107) 0.469† (0.283) 1.239*** (0.308) 1.181*** (0.276) 1.680*** (0.280) 0.295*** (0.046) 0.290*** (0.079) 0.487 (0.492) 0.009* (0.004) −0.745 (0.762) 0.231 36.670*** −0.933*** (0.261) −0.470** (0.170) −0.016** (0.005) 0.015* (0.006) 0.086† (0.051) 0.102 (0.126) 0.293† (0.176) 0.221* (0.113) 0.832*** (0.214) 0.136*** (0.033) −0.054 (0.034) −0.004 (0.255) 0.004* (0.002) 0.156 (0.331) −0.169 (0.258) 0.047 (0.239) −0.033*** (0.007) 0.006 (0.007) 0.264*** (0.059) 0.303† (0.172) 0.689*** (0.178) 0.93*** (0.172) 0.785*** (0.177) 0.121*** (0.027) 0.493*** (0.053) 0.351 (0.304) 0.002 (0.002) 0.841 (0.514) 0.108 906.110*** Note. Observation: 1,824. OLS = ordinary least squares; QA = quasi-governmental affiliation; GA = governmental affiliation; GDP = gross domain product. † p < .1. * p < .05. ** p<.01. *** p<.001. 20 Nonprofit and Voluntary Sector Quarterly 00(0) Figure 2. Interaction effects between nonprofit evaluation score and political affiliation with 95% confidence intervals. Note. The horizontal line represents civic foundations, the reference category. slightly reduced the probability of receiving charitable donations (see Models 2, 5, and 8). In double-hurdle models, board size and GDP per capita predicted whether a foundation would receive charitable donations but not the size of the donations. However, the two variables had similar marginal effects as in the OLS model (see Table 1). Nonprofit policy environment remained insignificant throughout the analysis. Discussion The sample of Chinese foundations and the research presented in this study indicate the extent to which organizational legitimacy attested through voluntary certification, Wang 21 Figure 3. Interaction effects between nonprofit tax exemption qualification and political affiliation with 95% confidence intervals. Note. The horizontal line represents civic foundations, the reference category. political connections, and their interactions are related to nonprofit fundraising under authoritarianism. The evidence shows that all the three certification programs, namely nonprofit evaluation, tax exemption qualification, and pretax deduction qualification can help foundations attract more charitable donations. This study distinguishes between individual and organizational political connections. Individual political connections or the number of government officials serving as managerial board members did not predict more charitable donations. There are two plausible reasons. First, while on the board, government officials had not developed a strong sense of identity and consequentially did not fulfill their expected function (Guo & Lai, 2019). Second, government officials were no longer as active in fundraising as before because the 22 Nonprofit and Voluntary Sector Quarterly 00(0) Figure 4. Interaction effects between nonprofit pretax deduction qualification and political affiliation with 95% confidence intervals. Note. The horizontal line represents civic foundations, the reference category. party-state had increasingly discouraged them from joining NGOs. Therefore, what (Xu, 2008) documented in her case study might have reduced in scale. Organizational political connections or affiliation was operationalized as a categorical variable. The separation of the two levels of political affiliation—quasi-governmental and governmental—offered nuances for their influence on fundraising. Although neither form of affiliation led to more charitable donations than civic foundations, they interacted with organizational legitimacy differently. In addition, if university foundation were removed from the regression, quasi-governmental foundations would have received more charitable donations than civic foundations. This is because most Chinese universities are PSUs in nature. They have a large physical presence 23 Wang Table 8. Average Marginal Effects of Political Affiliation on Charitable Donations (logged). Quasi-governmental = 2 Model variance covariance matrix of the estimators (VCE): Robust Civic foundations = 1 Nonprofit evaluation score 0 1 2 3 4 5 Nonprofit tax exemption 0 1 Nonprofit pretax deduction 0 1 Governmental = 3 OLS Double-hurdle OLS Double-hurdle dy/dx dy/dx dy/dx dy/dx (base outcome) (base outcome) (base outcome) (base outcome) 0.276 (0.280) 0.060 (0.263) −0.155 (0.282) −0.371 (0.330) −0.587 (0.398) −0.803† (0.477) 0.233 (0.277) −0.009 (2.257) −0.213 (0.301) −0.379 (0.376) −0.510 (0.460) −0.609 (0.547) 0.035 (0.311) −0.145 (0.290) −0.326 (0.302) −0.506 (0.341) −0.686† (0.402) −0.866† (0.475) −0.001 (0.304) −0.228 (0.283) −0.421 (0.298) −0.577† (0.335) −0.698† (0.384) −0.789† (0.443) 0.291 (0.299) −0.323 (0.345) 0.263 (0.293) −0.314 (0.389) 0.076 (0.316) −0.697† (0.400) 0.069 (0.307) −0.791* (0.392) 1.272*** (0.388) −0.432 (0.291) 1.188** (0.388) −0.493 (0.301) 0.487 (0.412) −0.404 (0.309) 0.402 (0.403) −0.411 (0.293) Note. Observation: 1,824. dy/dx for factor levels is the discrete change from the base level. VCE = variance covariance matrix of the estimators; OLS = ordinary least squares. † p < .1. * p < .05. ** p < .01. *** p < .001. visible to the public and through foundations promote their function as higher education providers over their political identity as PSUs. In this sense, quasi-governmental affiliation plays a subtle role. The above findings have wide implications for the authoritarian world. The first implication is about the nature of donations. States do not directly control charitable contributions. To help an NGO, governments need to utilize their authority to mobilize donations. Although states may privilege politically connected NGOs with fundraising advantage, they have been more successful in pressing certain individuals and organizations to donate than engaging a wide donor base. It is also possible that these NGOs 24 Nonprofit and Voluntary Sector Quarterly 00(0) have developed alternative sources of revenue, such as government subsidies and market returns (Ni & Zhan, 2017). The second implication relates to donors and incentives. Despite low transparency in many authoritarian states’ nonprofit sectors, donors may have become increasingly capable of perceiving NGO legitimacy. They do not necessarily embrace state-centered charity enthusiastically (Li et al., 2020) or assign legitimacy to NGOs’ political connections. Indeed, major scandals usually occurred in politically connected NGOs rather than in civic NGOs in regions such as Eastern Europe, Central Asia, Russia, and China (Smith et al., 2018). Furthermore, different from voluntary accreditation and certification programs administered by thirdparty watchdogs, governments can embed incentives in government-led programs through legislative and administrative requirements. This is especially true for the nonprofit pretax deduction qualification whose large coefficients might be a combined effect of perceived legitimacy and built-in incentives that cannot be disentangled with the current data. We may anticipate growing importance of this qualification because in 2018 the revised Enterprise Income Tax Law allowed corporate donations to be deductible up to 12% of firms’ profits and the excess to be carried forward to the same extent over the next 3 years. The interaction analysis provided evidence that under certain circumstances organizational legitimacy could negatively moderate the relationship between political affiliation and the amount of charitable donations. The three certification programs helped civic foundations raise more charitable donations than governmental and quasigovernmental foundations to varying degrees. Specifically, among foundations that were awarded an evaluation score of 1A or above or obtained either of the two tax qualifications, all the marginal effects except one in the OLS model were negative (see Table 8), and several were significant. The findings suggest the emergence of multiple, interactive patterns of fundraising that highlight the importance of organizational legitimacy and resource opportunities for civic foundations. The significant direct and indirect effect of organizational legitimacy led to the agreement with Cheng and Wu (2021, p. 56) that foundations were “able to survive and attract resources regardless of the status of their political connections.” However, in light of recent global experiences, we may be wise not to rush in assuming the dominance of organizational legitimacy in fundraising. Although the literature depicts a trend of closing civil society under authoritarianism (Heiss, 2019), more recent literature argues that the space for civil society may be changing or shifting rather than closing. The shrinking politically active part contrasts with the expanding service domain where loyal and neutral NGOs operate (Toepler et al., 2020). Since these NGOs do not impose an immediate threat, empowering them not only co-opts them into the social service apparatus but also institutionalizes regime resilience. Therefore, government-led accreditation and certification programs allow states to take advantage of welfare modernization, nonprofit professionalization, and charitable resources that the society is willing to contribute. Overall, the results indicate that the Chinese party-state has shifted its nonprofit policy from micro-management to macro-control. A greater number of disciplined Wang 25 civic NGOs would outweigh the loss of charitable resources that otherwise would have gone to politically connected NGOs because of the increase of total resources for charity. Nevertheless, these certification programs promoted standardization, which in turn encouraged nonprofit capacity-building, and provided benchmarks for the fundraising market. In this sense, they are beneficial for the nonprofit sector and the public’s understanding of NGOs under authoritarian rule. Conclusion The growth of NGOs intensified the competition for nonprofit resources under authoritarian rule. Our knowledge of NGOs’ resource acquisition, however, did not fully catch up with the new phenomenon in the nonprofit sector where they operate. This study is a timely examination of the dynamics between the state, nonprofits, and society to account for multiple patterns of nonprofit development in authoritarian states. Organizational legitimacy probably used to be irrelevant when almost all NGOs were created in a top-down manner. Nowadays, it can be responsible for NGOs’ donation income and moderating political connections in a way that enables civic foundations. By decoupling the separate and interactive effects of organizational legitimacy and political connections on charitable donations, this study offered novel perspectives to reconcile explanations for fundraising in China. Specifically, it expanded the scholarship on political connections through new variable operationalization and extended the literature on organizational legitimacy (i.e., voluntary accreditation and certification) and its moderating effect to the undertested authoritarian context. This study has four limitations. First, the data were cross-sectional, so it would be difficult to establish causality. Nor did the 1-year data allow this study to examine the effect of time lag (i.e., period a foundation has been certificated) and changes in certification results. Future research could gather longitudinal data and consider using interviews, surveys, and experiments to triangulate data. Surveys could include questions that separate certain voluntary certification programs to multiple analytical factors. Second, this article did not distinguish between institutional donors (i.e., firms) and individual donors. Future research may consider using regression methods on these two sources of donation separately on foundations’ legitimacy, political connections, and other organizational attributes to examine the extent to which these two types of donors would behave differently. Third, this study was not fully informed of the certification processes. Existing evidence seems to acknowledge the rigorousness of these certification programs.9 Future research may investigate the experience of NGOs that are certified or seeking certification to test the extent to which these programs truly corroborate legitimacy. Finally, although contemporary authoritarian regimes have exhibited similar nonprofit policy, they differ in economic and social specifics. Therefore, we should adaptively interpret the findings and define organizational legitimacy and political connections for each authoritarian state in a way that is appropriate for its institutional reality. 26 Nonprofit and Voluntary Sector Quarterly 00(0) Acknowledgments I thank Dr. Beth A. Gazley, Dr. James L. Perry, Dr. Jennifer N. Brass, Dr. Jason Y. Wu, Dr. Chao Guo, Dr. Nicolas J. Duquette, Dr. Seth M. Freedman, and other colleagues for their constructive comments. I also thank the three anonymous reviewers who provided constructive suggestions. Declaration of Conflicting Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. Funding The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study received funding from the Beijing Yifang Foundation and the Zhejiang Dunhe Foundation. Data Availability The instructions, data, and codes that produce the findings reported in this article are available at http://dx.doi.org/10.13140/RG.2.2.23785.98400. ORCID iD Qun Wang https://orcid.org/0000-0001-7592-9824 Supplemental Material Supplemental material for this article is available online. Notes 1. 2. 3. 4. 5. 6. 7. 8. See Russia at https://www.themoscowtimes.com/2018/12/13/russians-donate-5-7-billiondollars-to-charities-per-year-report-says-a63810; Turkey at https://www.tusev.org.tr/usrfiles/images/belgeler/Individual_Giving_and_Philanthropy_in_Turkey.pdf; and China in Table 2. See http://www.chinanews.com/gn/2013/09-22/5305391.shtml. See an example at http://wemedia.ifeng.com/27476671/wemedia.shtml. The Eleventh Chinese Private Enterprise Survey showed that in 2012–2013, Chinese private firms donated to both politically connected and unconnected NGOs. For more information, see https://peng.baiji.org.cn/?p=772. See at http://finance.huanqiu.com/people/2012-07/2894862.html?agt=61. See RICF at https://github.com/ma-ji/RICF; Wang (2018a) at http://doi.org/10.13140/ RG.2.2.35141.55525; and Wang (2018b) at https://static-content.springer.com/esm/art%3 A10.1007%2Fs41111-018-0094-2/MediaObjects/41111_2018_94_MOESM1_ESM.xlsx. A reviewer commented that public fundraising qualification should be a proxy of organizational legitimacy variable, too. This could be so after 2016, but public fundraising qualification was not a certified qualification in 2013. See a discussion on public fundraising qualification and analysis of this variable in the Supplemental Online Material. I thank the reviewer who suggested adding university foundation as a control variable. Wang 9. 27 See examples at http://www.chinadevelopmentbrief.org.cn/news-22152.html and http:// www.bnu1.org/show_1960.html. References Adena, M., Alizade, J., Bohner, F., Harke, J., & Mesters, F. (2019). Quality certification for nonprofits, charitable giving, and donors’ trust: Experimental evidence. Journal of Economic Behavior and Organization, 159, 75–100. Bekkers, R. (2003). Trust, accreditation, and philanthropy in the Netherlands. Nonprofit and Voluntary Sector Quarterly, 32(4), 596–615. Brody, E. (2002). Accountability and public trust. In L. M. Salamon (Ed.), The state of nonprofit America (pp. 471–498). Brookings Institution. Cheng, Y., & Wu, Z. (2021). The contingent value of political connections on donations to Chinese foundations: Exploring the moderating role of transparency. Administration & Society, 53(1), 36–63. Feng, L., & Zhang, Y. (2014). The field of philanthropy: The duo-variation of administration and de-administration. Rutgers Huamin Research Center. https://pdfs.semanticscholar.org /6893/05fdecd33843102b78383ce809633e26eca6.pdf Feng, N. C., Nelly, D. G., & Slatten, L. A. D. (2016). Accountability standards for nonprofit organizations: Do organizations benefit from certification programs? International Journal of Public Administration, 39(6), 470–479. Feng, N. C., Nelly, D. G., & Slatten, L. A. D. (2019). Stakeholder groups and accountability accreditation of non-profit organizations. Journal of Public Budgeting, Accounting & Financial Management, 31(2), 218–236. Fischer, R. L., Wilsker, A., & Young, D. R. (2011). Exploring the revenue mix of nonprofit organizations: Does it relate to publicness? Nonprofit and Voluntary Sector Quarterly, 40(4), 662–681. Forbes, K. F., & Zampelli, E. M. (2011). An assessment of alternative structural models of philanthropic behavior. Nonprofit and Voluntary Sector Quarterly, 40(6), 1148–1167. Galaskiewicz, J., & Bielefeld, W. (1998). Nonprofit organizations in an age of uncertainty. Aldine de Gruyter. Gibelman, M., & Gelman, S. R. (2004). A loss of credibility: Patterns of wrongdoing among nongovernmental organizations. VOLUNTAS: International Journal of Voluntary and Nonprofit Organizations, 15(4), 355–381. Gugerty, M. K. (2009). Signaling virtue: Voluntary accountability programs among nonprofit organizations. Policy Sciences, 42(3), 243–273. Guo, C. (2007). When government becomes the principal philanthropist: The effects of public funding on patterns of nonprofit governance. Public Administration Review, 67(3), 458– 473. Guo, C., & Lai, W. (2019). Community foundations in China: In search of identity? VOLUNTAS: International Journal of Voluntary and Nonprofit Organizations, 30(4), 647–663. Heiss, A. (2019). NGOs and authoritarianism. In T. Davies (Ed.), Routledge handbook of NGOs and international relations (pp. 557–572). Routledge. Henderson, S. (2011). Civil society in Russia: State-society relations in the post-Yeltsin era. Problems of Post-communism, 58(3), 11–27. Ho, P. (2007). Embedded activism and political change in a semiauthoritarian context. China Information, 21(2), 187–209. Hsu, C. L., & Jiang, Y. (2015). An institutional approach to Chinese NGOs: State alliance versus state avoidance resource strategies. The China Quarterly, 221, 100–122. 28 Nonprofit and Voluntary Sector Quarterly 00(0) Hsu, J., & Hasmath, R. (2017). A maturing civil society in China? The role of knowledge and professionalization in the development of NGOs. China Information, 31(1), 22–42. Hu, M., & Kong, D. (2021). Examining Chinese charitable giving under state intervention: A political participation perspective. Nonprofit and Voluntary Sector Quarterly, 1–24. https:// doi.org/10.1177/08997640211057438 Hwang, H., & Powell, W. W. (2009). The rationalization of charity: The influences of professionalism in the nonprofit sector. Administrative Science Quarterly, 54(2), 268–298. Jia, X. (2011). Civil organization-government relationships: Functional cooperation and power dilemmas. In M. Wang (Ed.), Emerging civil society in China, 1978–2008 (pp. 209–237). Brill. Johnson, J. G., & Ni, N. (2015). The impact of political connections on donations to Chinese NGOs. International Public Management Journal, 18(4), 514–535. Kang, X. (2018). Moving toward neo-totalitarianism: A political-sociological analysis of the evolution of administrative absorption of society in China. Nonprofit Policy Forum, 9(1), 1–8. Li, H., Lo, C. W., & Tang, S. (2017). Nonprofit policy advocacy under authoritarianism. Public Administration Review, 77(1), 103–117. Li, H., Tang, S. Y., & Lo, C. W. H. (2018). The institutional antecedents of managerial networking in Chinese environmental NGOs. Nonprofit and Voluntary Sector Quarterly, 47(2), 325–346. Li, M., Tao, R., & Su, F. (2020). Bringing politics back in charitable giving: Evidence from donations after China’s Sichuan earthquake. Nonprofit and Voluntary Sector Quarterly, 49(2), 272–291. Livshin, A., & Weitz, R. (2006). Civil society and philanthropy under Putin. International Journal of Not-for-Profit Law, 8(3), 7–12. Ma, J., Wang, Q., Dong, C., & Li, H. (2017). The research infrastructure of Chinese foundations, a database for Chinese civil society studies. Scientific Data, 4, 170094. Ma, Q. (2006). Non-governmental organizations in contemporary China: Paving the way to civil society?. Routledge. Newman, C., Henchion, M., & Matthews, A. (2003). A double-hurdle model of Irish household expenditure on prepared meals. Applied Economics, 35(9), 1053–1061. Nguyen, P. A., & Doan, D. R. H. (2015). Giving in Vietnam: A nascent third sector with potential for growth. In P. Wiepking & F. Handy (Eds.), The Palgrave handbook of global philanthropy (pp. 473–487). Palgrave Macmillan. Ni, N., & Zhan, X. (2017). Embedded government control and nonprofit revenue growth. Public Administration Review, 77(5), 730–742. Ortmann, A., & Myslivecek, J. (2010). Certification and self-regulation of nonprofits, and the institutional choice between them. In B. A. Seaman & D. R. Young (Eds.), Handbook of research on nonprofit economics and management (pp. 280–290). Edward Elgar Publishing. Parsons, L. M. (2007). The impact of financial information and voluntary disclosures on contributions to not-for-profit organizations. Behavioral Research in Accounting, 19(1), 179– 196. Pearson, M. M. (2007). Governing the Chinese economy: Regulatory reform in the service of the state. Public Administration Review, 67(4), 718–730. Peng, J. (2014). Zhongguo cishan touming baogao (2009–2014) [China philanthropic transparency report (2009–2014)]. Enterprise Management Publishing House. Peregudov, S., & Semenenko, I. (1996). Lobbying business interests in Russia. Democratization, 3(2), 115–139. Powell, W. W., & DiMaggio, P. J. (Eds.). (1991). The new institutionalism in organizational analysis. University of Chicago Press. Wang 29 Princeton Survey Research Associates. (2001). BBB Wise Giving Alliance donor expectations survey. Richter, J., & Hatch, W. (2013). Organizing civil society in Russia and China: A comparative approach. International Journal of Politics, Culture, and Society, 26(4), 323–347. Saxton, G. D., & Wang, L. (2014). The social network effect: The determinants of giving through social media. Nonprofit and Voluntary Sector Quarterly, 43(5), 850–868. Slatten, L. A. D., Guidry, B. N., & Austin, W. (2011). Accreditation and certification in the non-profit sector: Organizational and economic implications. Organization Management Journal, 8, 112–127. Smith, D. H., Moldavanova, A., & Krasynska, S. (Eds.). (2018). The nonprofit sector in Eastern Europe, Russia, and Central Asia. Brill. Suchman, M. C. (1995). Managing legitimacy: Strategic and institutional approaches. Academy of Management Review, 20(3), 571–610. Toepler, S., & Fröhlich, C. (2020). Advocacy in authoritarian contexts: The case of disability NGOs in Russia. International Journal of Sociology and Social Policy, 40(11/12), 1473–1489. Toepler, S., Zimmer, A., Fröhlich, C., & Obuch, K. (2020). The changing space for NGOs: Civil society in authoritarian and hybrid regimes. VOLUNTAS: International Journal of Voluntary and Nonprofit Organizations, 31, 649–662. Wang, H., & Qian, C. (2011). Corporate philanthropy and corporate financial performance: The roles of stakeholder response and political access. Academy of Management Journal, 54(6), 1159–1181. Wang, Q. (2018a). Have foundations become an independent sector in China? An examination of the links between foundations and the state. Asia Pacific Journal of Public Administration, 40(1), 68–73. Wang, Q. (2018b). A typological study of the recent development and landscape of foundations in China. Chinese Political Science Review, 3(3), 297–321. Wang, Q. (2020). Government control, governance and spatial distribution: What we have learned from the Research Infrastructure of Chinese Foundations. Asia Pacific Journal of Public Administration, 42(1), 33–45. Wang, Q. (2021a). The development and operation of foundations in China. The Foundation Review, 13(1), 36–56. Wang, Q. (2021b). Multilevel political connections and nonprofit revenues under authoritarianism: Evidence from Chinese Foundations. Governance. Advance online publication. https://doi.org/10.1111/gove.12663 Wang, Q., & He, L. (2018). Are the wealthy also healthy? An empirical evaluation of the financial health of Chinese foundations. Chinese Public Administration Review, 9(1), 6–24. Wang, Z., Gao, J., & Liu, Y. (2016). Zhongguo shehui zhengce jinbu zhishu baogao [China social policy progress index report]. Social Sciences Academic Press. Wei, Q. (2020). From direct involvement to indirect control? A multilevel analysis of factors influencing Chinese foundations’ capacity for resource mobilization. VOLUNTAS: International Journal of Voluntary and Nonprofit Organizations, 31, 762–778. Wiepking, P., & Handy, F. (2015). Explanations for cross-national differences in philanthropy. In P. Wiepking & F. Handy (Eds.), The Palgrave handbook of global philanthropy (pp. 9–21). Palgrave Macmillan. Xu, Y. (2008). Feiduichen xing yilai: Zhongguo jijinhui yu zhengfu guanxi yanjiu [Unsymmetrical dependence: Study on the relationship between foundations and government in China]. Journal of Public Management, 5(1), 33–40. 30 Nonprofit and Voluntary Sector Quarterly 00(0) Zhang, J., Marquis, C., & Qiao, K. (2016). Do political connections buffer firms from or bind firms to the government? A study of corporate charitable donations of Chinese firms. Organization Science, 27(5), 1307–1324. Zhao, R., Wu, Z., & Tao, C. (2016). Understanding service contracting and its impact on NGO development in China. VOLUNTAS: International Journal of Voluntary and Nonprofit Organizations, 27(5), 2229–2251. Zhou, H., & Ye, S. (2021). Fundraising in the digital era: Legitimacy, social network, and political ties matter in China. VOLUNTAS: International Journal of Voluntary and Nonprofit Organizations, 32, 498–511. Author Biography Qun Wang is an assistant professor in the Department of Political Science and Public Administration at the University of Toledo. He holds a PhD in public policy from Indiana University Bloomington. His research interests include government–nonprofit relations and social innovation.