The Myth of Chinese Think Tanks: Determinants of Chinese Think Tanks' Advocacy Role

(1)

The Myth of Chinese Think Tanks

Determinants of Chinese Think Tanks’ Advocacy Role

Master Thesis, 11

th

of June 2018

MSc Public Administration, Leiden University

Author: Danqi Feng (s2031361)

Supervisor: Dr. Valérie Pattyn

(2)

Determinants of Chinese Think Tanks’ Advocacy Role

Table of content

1. Introduction ... 3

1.1 Research question ... 3

1.2 Scientific and practical relevance ... 7

1.3 Thesis structure ... 8

2. Literature Review and Theoretical Framework ... 9

2.1 Chinese think tanks: definition ... 9

2.2 Chinese think tanks: classification ... 13

2.3 The difference between Chinese think tanks: administrative linkages ... 19

2.4 Chinese think tanks: roles and activities ... 21

2.5 Possible determinants of the extent of advocacy activities of Chinese think tanks .. 24

2.6 Theoretical framework and summary of hypotheses ... 28

3. Research Design and Data Collection ... 30

3.1 Operationalization and measurement of key variables ... 30

3.1.1 Independent variables ... 30

3.1.2 Dependent variable ... 31

3.2 Research design ... 33

3.3 Data collection ... 35

4. Results and Analysis ... 37

4.1 Descriptive statistics ... 37

4.2 Correlations ... 41

4.3 Regression analysis ... 43

4.3.1 Administrative linkages and the extent of advocacy activities ... 43

4.3.2 Regional knowledge capacity and the extent of advocacy activities ... 45

4.3.3 Type of policy areas and the extent of advocacy activities ... 46

4.3.4 Multivariate regression: the effects of regional knowledge capacity and policy

areas on the extent of advocacy activities ... 47

5. Discussion and Conclusion ... 52

5.1 Theoretical explanations ... 52

5.2 Result of interview ... 55

5.3 Strengths and limitations ... 57

5.4 Conclusion ... 59

References ... 62

Appendix 1 ... 66

(3)

1. Introduction

There is an increasing interest in understanding Chinese think tanks, as Chinese think tanks have grown rapidly both in quantity and quality in the past decade. According to a report by the Think Tanks and Civil Societies Program (TTCSP) of the Lauder Institute, Chinese Mainland is hosting the second largest number of think tanks in the world with 512 think tanks (McGann, 2018). Not only that, but the TTCSP has rated 9 Chinese think tanks as the world’s 173 top think tanks (McGann, 2018) in terms of the quality of think tanks’ research and their impact on policymaking. Correspondingly, scholars have been motivated to investigate Chinese think tanks (e.g.: Stone, 2005; Nachiappan, 2013).

1.1 Research question

Existing knowledge of Chinese think tanks is massively based on US experience (Hayward, 2018; Köllner, Zhu & Abb, 2018). This is not surprising, given that think tanks originate from America and America is hosting the largest number of think tanks in the world (McGann, 2018). This however may prevent a further exploration and understanding of Chinese think tanks and their roles in policymaking process. Based on US experience, Western scholars have criticised China for not having the ‘true’ think tanks (Hayward, 2018). One common premise about think tanks is that they need to be independent of the government and political parties in their operation and funding (McGann, 2018). However, Chinese think tanks cannot be entirely independent of the government and political parties due to the one-party regime. The Communist Party of China (CPC) is considered to be strong because public officials in Chinese government highly overlap with the CPC, and Party Branches of the CPC have permeated through Chinese organisations including enterprises, education institutes, research institutes and other grass-roots units. Chinese think tanks in this context are more or less related to the government and the CPC, whether officially or privately. Using independence as a criterion, most Western scholars thus state that China does not possess the ‘true’ think tanks. Moreover, because of a strict control of speech in China by the CPC and the government, Western scholars commonly see Chinese think tanks as organizations that speak for the state instead of venturing the interest of society (Hayward, 2018; Popović, 2017). They insist that Chinese think tanks dare not to advocate their own positions and policy ideas, even though their policy advice is founded on scientific knowledge,

(4)

because advocating their policy ideas might lead to serious consequences, especially when think tanks’ positions conflict with that of policy makers (Popović, 2017). In this sense, scholars regard Chinese think tanks as mere advisors to policy makers: they are tools that the Chinese government and political parties use to justify and support the ideas of policy makers, rather than enablers that help policy makers to make informed decisions.

Scholars have made an endeavour to uncover Chinese think tanks’ role in such a unique host society. The existing studies of Chinese think tanks are mainly centred in three research topics. One is the definition and classification of Chinese think tanks. Scholars, such as Xue & Zhu (2006), try to develop a concept and taxonomy of think tanks in the Chinese context. According to them, the criterion of independence is not suitable for Chinese think tanks considering that most Chinese think tanks are associated with the government in their funding, personnel or/and operation. The use of research autonomy could be a solution to understanding Chinese think tanks. Autonomy, as defined by them, refers to the freedom of deciding research topic by think tanks (Xue & Zhu, 2006). Another heated topic is the measurement of Chinese think tanks’ influence. Scholars focusing on this topic attempt to construct a set of metrics to assess Chinese think tanks’ impact on policymaking process (e.g.: Chen & Meng, 2015). The last topic is the development (rise or fall) of Chinese think tanks. For example, Köllner et al. (2018) explore the driving forces of Chinese think tanks’ development and challenges that Chinese think tanks may encounter.

Regardless of all these studies, little is known about the role that Chinese think tanks fulfil, activities that they engage in, and determinants of the type of activities they are involved in. More in particular, there is hardly any evidence about the role that Chinese think tanks engage in: do Chinese think tanks merely function as advisors to policy makers or do they promote themselves as well, what kind of advocacy activities do Chinese think tanks engage in, and what affects the extent of Chinese think tanks’ advocacy activities? As the above discussion suggests, existing literature on Chinese think tank is primarily based on American experience, which leads to a common skepticism among Western academia about Chinese think tanks’ role in the public policymaking process. More particularly, Western scholars doubt whether Chinese think tanks engage in advocacy activities given

(5)

the specific Chinese context.

Zhu’s (2011) analytical framework of Chinese think tanks’ role provides insights into the question whether Chinese think tanks can function as advocates. According to him, Chinese think tanks simultaneously act as “advisors to the government, academics in universities and research institutes, and policy advocates in the public sphere” in the policymaking process (Zhu, 2011, p. 672). Different role involves diverse types of activities. By analysing the type of activities think tanks are involved in, one can identify the role that think tanks are taking. Playing the role of advocates entails advocacy activities that target at the public. This type of activities is characterized by the use of media and internet, and it has a goal of indirectly influencing policy makers by appealing for public support for think tanks’ positions and policy ideas. According to Zhu’s (2011) observation, in addition to advisory activities, Chinese think tanks put numerous efforts into advocacy activities. Similarly, Popović’s (2017) study of Chinese citizens’ groups gives a detailed account of advocacy activities by Chinese interest groups. By conducting 38 interviews with Chinese citizens’ groups, Popović (2017) reveals that even though Chinese citizens’ groups exchange their sources for direct access to policy makers, they do utilise diverse strategies to mobilise the public and to obtain public support for their policy positions. Moreover, those interviews suggest that in the eyes of Chinese citizens’ groups, advocating their policy positions is even more important than directly approaching policy makers. In contrast to what most Western scholars have presumed, Chinese interest groups do engage in advocacy activities to influence policymaking process regardless of the strict speech control by the Chinese government and the CPC. Think tanks can be regarded as a type of interest groups in the sense that they have the goal of influencing policymaking through scientific knowledge. Therefore, this study contends that Chinese think tanks also play the role of advocates in the public domain in addition to the role of advisors.

Even though Chinese think tanks do play the role of advocates, this study argues that they differ in the extent of advocacy activities. Regarding determinants of this difference, a body of literature provides diverse insights. The insider/outsider theory by Grant (2004) suggests that a direct access to policy makers, or in terms of “administrative linkage” (Zhu, 2011, p. 341), can be seen as a

(6)

determinant of this difference. Grant (2004) claims that the direct access to policy makers affects interest groups’ preference for strategies of influencing the policymaking process. For insider groups that have direct access to policy makers, they preferably utilise insider strategies to directly influence policymaking, such as contacting policy makers. While for outsider groups without privileged access to policy makers, they more frequently resort to outsider strategies to mobilise the public, such as contacting the press and holding public meetings. Based on this theory, one can argue that the direct access to policy makers/administrative linkages can determine whether Chinese think tanks primarily engage in advocacy activities.

Radaelli’s (1999) theory of expertise in politics tells a different story. In light of this theory, it is assumed that think tanks’ policy areas determine the extent of advocacy activities of think tanks. Radaelli (1999) argues that ‘salience’ and ‘uncertainty’ of policy issues are two dimensions that together decide whether politicians will adopt expertise. Salience refers to the extent of a policy issue being visible to the public and being influenced by media and the public, while uncertainty indicates whether the policy issue is complex and demands a high level of information and expertise (Radaelli, 1999). In an uncertain policy area (e.g.: aviation), expertise is most likely to be adopted by politicians as dealing with such kind of policy issues demands a great deal of knowledge. In a salient policy area (e.g.: monetary provisions), the public is most likely to be mobilised since policy issues are open to media and public discussion. Following this, it can be asserted that think tanks that active in salient policy areas prefer more advocacy activities; they more often take advantage of media and the public to get their policy ideas heard by policy makers.

Another argument about the determinants of the extent of advocacy activities of Chinese think tanks can be found in Zhu’s (2011) work, which concludes that regions where think tanks are located in can determine the extent of think tanks’ advocacy activities. Chinese regions vary in their knowledge capacity for procuring, taking in, and disseminating knowledge (Zhu, 2011). In a high knowledge capacity region, such as Beijing city, think tanks are faced with less resistance when promoting their policy ideas, because residents there are more willing to embrace and absorb new ideas. Consequently, think tanks in that regions engage in more advocacy activities to obtain public

(7)

support. In contrast, think tanks located in low knowledge capacity regions, such as Henan province, experience more difficulty in disseminating their policy ideas, as residents are reluctant to take in new knowledge. In this case, engaging in advocacy activities is less effective and efficient to influence policymaking compared to directly approaching policy makers. Think tanks therefore prefer advisory activities to advocacy activities.

This thesis aims to shed light on this ambiguity, by systematically examining the several potential explanatory factors that help to understand the extent of Chinese think tanks’ advocacy activities. The key research question for this study is:

RQ: Which determinants explain the extent of advocacy activities Chinese think tanks engage in?

In addition, this study focuses on Chinese semi-official think tanks. The reason for this focus is that investigating advocacy activities of Chinese semi-official think tanks can directly respond to the criticism that China does not possess the ‘true’ think tanks. As previously introduced, because of a lack of independence of the government and political parties, western scholars commonly regard Chinese think tanks as mere advisors to policy makers rather than advocates. Following this line of reasoning, Chinese semi-official think tanks are the least likely one to play the role of advocates in the policymaking process since they are by name more closely related to the government than civil think tanks. By looking at the content and extent of advocacy activities Chinese semi-official think tanks engage in, this study is able to respond to the critique by western scholars.

1.2 Scientific and practical relevance

This study builds upon and aims to contribute to think tank literature and the broader field of public administration in two ways. Firstly, within the public administration domain, scholars have made a great number of efforts to research into the role of scientific knowledge in the public policymaking process. Still, there is an ongoing argument about this. Some scholars have embraced research-based/evidence-based public policymaking, while others have begun questioning the authority and contestability of scientific knowledge in public policymaking (Hart & Vromen, 2008).

(8)

One of the reasons for this is that knowledge more often serves politicians as a justification for policy makers’ decisions rather than as the scientific evidence of public policy. This study seeks to shed light on this argument by researching the role of think tanks in policy making process, organisations that are exclusively devoted to producing research-based policy analysis and influencing policymaking. Through analyzing and measuring the advocacy activities Chinese think tanks are involved in, this study provides an empirical answer to the question that whether think tanks promote scientific knowledge to the public.

Secondly, within the area of think tanks, despite the increasing interest in and growing studies of Chinese think tanks, it is acknowledged that a large amount of literature is centred in America and Europe (Köllner et al., 2018). Literature on Chinese think tanks is under-developed compared to that of English-speaking countries. In addition, as regards Chinese think tanks’ advocacy role and determinants of this role, there is no literature that exclusively targets at this issue. The aim of this study is thus to fill in the academic gap of Chinese think tanks and to answer the un-researched question of Chinese think tanks’ advocacy role in policymaking process and determinants of this role.

Practical relevance of this study is twofold. On the one hand, analysing and assessing the advocacy activities of Chinese think tanks helps to delineate the role that they play in public policymaking process. This further clarifies the ambiguity of whether Chinese think tanks can be advocates in the public domain. On the other hand, by giving an account of under what circumstances Chinese think tanks primarily engage in what type of advocacy activities, this study can provide Chinese think tanks with a guide to choosing advocacy strategies.

1.3 Thesis structure

This study is divided in five sections. In the next section, Secion 2, the relevant literature and theories are reviewed. This section also develops four sets of hypotheses and presents the theoretical framework for this study. In section 3, the research design and data collection are outlined. Issues regarding operationalization, sample selection, method design, and data collection are addressed in details. Section 4 presents the data processing and data analysis. Section 5 provides a final conclusion,

(9)

limitations of the current study. At the end, the Reference section and the Appendix section provide information about the relevant literature and interview document.

2. Literature Review and Theoretical Framework

This section elaborates on the theoretical framework that is used to shed light on the research question of this study. The first part of this section reviews literature on definition of think tanks, and clarifies the definition upon which this study builds. The next part examines the current studies about think tanks’ classification, based on which this study develops a categorization of Chinese think tanks. A more detailed discussion of the difference between diverse types of Chinese think tanks is given as a subsequent part. The succeeding part reviews theories regarding think tanks’ role, followed by a discussion of possible determinants of think tanks’ roles. Hypotheses are subsequently formulated based on these theories; they together represent the proposed answers to the research question.

2.1 Chinese think tanks: definition

Think tanks worldwide differ significantly in size, policy areas and significance, which makes it difficult to have a unified concept of think tanks (Stone, 2003). Scholars diverge at the definition of think tanks. Think tanks in the widest sense can be understood as “institutions that provide public policy research, analysis, and advice” (McGann & Weaver, 2000, p. 4, cited in McGann, 2016, p. 5). This definition reflects that the fundamental function of think tanks is to engage in public policy through the means of research. However, this definition can be applied to too many types of organisations, such as interest groups and government agencies. To narrow it down, scholars define think tanks as public policy research organisations that are independent of the government, political parties and interest groups in their funding and operations. They regard ‘being independent of the government, political parties and interest groups’ as a tenet of think tanks (Stone, 2005; McGann, 2016).

The emphasis of independence sparks controversy. For example, McGann (2000) claims that independence is not a criterion when it comes to think tanks’ definition. According to him, think tanks can be both affiliated with the government, or independent nongovernmental organisations (NGOs)

(10)

(McGann, 2000). What differentiates think tanks from other organisations is that think tanks are permanent institutions rather than ad hoc (McGann, 2000; McGann, 2016). Meanwhile, McGann (2000) endorses that think tanks need to represent the voice of the public interest and to influence public policy by connecting scientific knowledge and public policy. Additionally, Stone and Ullrich (2003, p. 5) also criticise that the ‘independence’ premise is an “Anglo-American understanding of think tanks”, which cannot be directly applied to other political cultures, such as the European systems. In consideration of the complex multi-level EU governance system, Stone and Ullrich (2003) suggest that defining European think tanks’ role in the policy process is more critical to understand European think tanks than measuring European think tanks’ independence. They further propose that three dimensions of independence can be used to discuss European think tanks: “legal independence”, “financial independence”, and “scholarly independence” (Stone & Ullrich, 2003, p. 6).

Likewise, the understanding of Chinese think tanks cannot simply be copied from the Anglo-American model. The principle of ‘being independent of the government, political parties and interest groups’ is not suitable for Chinese think tanks, as Chinese think tanks are more or less related to the government and the CPC in their organizational operation and funding. On the one hand, the CPC is regarded to be strong because of the one-party regime in China. Party Branches of the CPC are all over the Chinese organisations, including companies, education institutes and other types of organisations. They are responsible for carrying out the Party's principles and policies, providing guidance to and supervising organisations in obeying the laws and regulations of the state. In such a context, the tenet of being independent is no longer applicable to Chinese think tanks. On the other hand, most Chinese think tanks are financially dependent on or related to the government. As Stone (2005) observes, Chinese think tanks are often established or/and funded by the government, and researchers “often work in patron-client relations with political leaders” (p. 3). In light of this situation, Stone (2005) further suggests that think tanks in the Asian political culture is more appropriately conceived as a function of conducting policy research and providing policy advice, rather than a “specific legal organizational structure as a non-governmental, non-partisan or independent civil society entity.” (p. 3). This viewpoint demonstrates that independence is not a one-size-fits-all principle of think tanks when political culture is taken into account. However, Stone’s

(11)

(2005) definition of Asian think tanks is too ambiguous, as it does not delineate what think tanks essentially are. For example, it is hard to differentiate think tanks from universities.

There is a considerable number of studies exploring the definition of think tanks in the Chinese context. One ongoing debate is that what independence means for Chinese think tanks. An interview in 2016 with Hu Angang—head of the Institute for Contemporary China Studies—by Hayward (2018) reveals that independence of think tanks in the Chinese context can be interpreted as three factors. One is that think tanks are able to select research topic autonomously (Hayward, 2018). That is to say, think tanks themselves can decide what topic they will be engaged in. Another one is that think tanks are able to conduct research autonomously (Hayward, 2018). This means think tanks can conduct research without interference from other actors, such as the government, corporates, and interest groups. The last one is that think tanks can publish their results independently (Hayward, 2018). Similarly, Zhu (2009) claims that research autonomy in deciding research topic can be a criterion to define Chinese think tanks. Both Hu’s and Zhu’s (2009) take into account the Chinese one-party regime, however, they do not illustrate how to measure those three factors or research autonomy. By reviewing western and Chinese literature on think tanks, Xue and Zhu (2009) develop their definition of Chinese think tanks: Chinese think tanks are stable and autonomous organisations that do policy research and policy advisory practices. According to them, think tanks are stable in the sense that they are permanent organisations. On the other hand, think tanks need to be administratively independent of any government agency and the CPC, in the sense that they possess an independent legal status. This definition by Xue and Zhu (2009) avoids the pitfall of think tanks’ independence, and it clarifies clearly how to measure think tanks’ interdependence in the Chinese context: by looking at their independent legal statuses. An overview table of think tanks’ definitions is provided (Table 1).

(12)

Author(s) Definition of think tanks Assessment

McGann & Weaver (2000)

Think tanks are institutions that provide public policy research, analysis, and advice.

1. It is the broadest understanding of think tanks.

2. It qualifies interest groups and government agencies as well.

Stone (2001)

Think tanks are public policy research organisations that are independent of the government, political parties and interest groups.

1. Anglo-American understanding of think tanks: independence is important.

2. It cannot be simply applied to other political cultures.

McGann (2016)

Think tanks are independent permanent research institutions that conduct public policy research and provide policy advice on domestic and international issues. They can be affiliated with political parties, the government, or universities.

1. It emphasizes on think tanks’ nature of being

permanent.

Definition of Chinese think tanks

Stone (2005)

Think tank in Asian culture is more a function of conducting policy analysis and providing policy advice, than a specific legal organizational structure.

1. It is too ambiguous in the sense that it qualifies too many types of organisations.

Hu Angang, cited from Hayward

(2018)

Zhu (2009)

Think tanks are policy research organisations that are able to select research topic autonomously, to conduct research autonomously, and to publish their results

independently.

1. Independence is interpreted into three

factors: autonomy in

selecting topic, conducting research, and publishing results. 2. Factors are hard to

clarify.

Xue & Zhu (2009)

Think tanks are permanent and legally independent organisations that do policy research and advisory practices.

1. Independence is translated into an

independent legal status.

(13)

In spite of those diverse definitions of think tanks, three common elements in these definitions can be identified. Firstly, think tanks are essentially research organisations. One of their fundamental functions is to conduct policy research and analysis and provide policy advice. Secondly, it can be said that one of the primary goals of think tanks is to influence policymaking. They do so by translating their research outcomes into products (e.g.: articles, reports) that are accessible and understandable to government, political parties, and the public. They are recognized as a bridge between research and policy. Thirdly, think tanks need to be independent, even though independence can be interpreted differently in diverse political cultures. The Anglo-American tradition conceives independence as think tanks being completely unattached to the government, political parties and interest groups (Stone 2005). Independence in the European system is construed in three dimensions, namely the legal independence, financial independence, and scholarly independence (Stone & Ullrich, 2003). Independence in the Chinese context can be either translated into research autonomy in selecting research topic, conducting research, and publishing results (Hayward, 2018; Zhu, 2009), or as an independent legal status (Xue & Zhu, 2009).

All put together, this study defines Chinese think tank as research organisations that conduct policy research and provide policy advice as an independent legal personality; they are operated with the goal of influencing policymaking.

2.2 Chinese think tanks: classification

Scholars classify think tanks in different ways. Weaver (1989) breaks think tanks into three types. One is “universities without students” (p. 564). These think tanks heavily rely on long-term academic research. They are more frequently funded by private sectors and they are devoted to book-length research outcomes. Another is “contract research organisations” (Weaver, 1989, p. 566). Their research agenda is set by agencies that contract with and pay for them. In contrast, the research products of contract research organisations are usually in the form of reports that specifically target at government agencies. The last one is “advocacy tanks” (Weaver, 1989, p. 567). Advocacy tanks have a strong focus of policy or ideology and a determination to influence policy debates. They more frequently “synthesize…[the] existing research rather than carrying out original research” (Weaver, 1989, p. 567). The taxonomy of think tanks by Weaver (1989) carries several disadvantages. Firstly,

(14)

the focus on book-length research products of universities without students makes it difficult or even impossible to influence policymaking, as policy makers barely have time to read book-length products produced by academic researchers. This type of research products is neither practical for policy makers nor understandable to the public. Secondly, it is difficult to unambitiously apply this taxonomy to empirical practice. In fact, many think tanks today possess multiple characteristics, for example, they can be universities without students as well as contract research organisations at the same time.

McGann (2000; 2016) develops a categorization of think tanks by distinguishing between independent and affiliated think tanks. Independent think tanks are intellectually, financially, and legally free from the influence of the government, corporates, and interest groups. They are able to conduct research and advocate their positions without any interference. This independence is considered to be critical to think tanks’ credibility (McGann, 2000). Following this, McGann (2000) proposes that there are four types of independent think tanks that can be found in America, namely the academic-diversified and academic-specialized think tanks, contract research organisations, advocacy think tanks, and the policy enterprise organisations. Meanwhile, McGann (2000) admits that think tanks in other regions have been traditionally associated with the government, political parties, and corporates, whether in an administrative, financial, or/and legal way (e.g.: European think tanks, Asian think tanks). Therefore, he advances four types of affiliated think tanks based on the type of affiliation, namely the party-affiliated think tanks, government sponsored think tanks, private think tanks (for profit), and university-based think tanks. McGann (2000; 2016) takes into account other political cultures when classifying think tanks. Nevertheless, this taxonomy cannot be applied to Chinese think tanks. The category of independent think tanks is not suitable for Chinese think tanks, because Chinese think tanks are not able to be entirely independent of the government and the CPC. The category of affiliated think tanks contradicts the foregoing discussion in the first part of this section. This study defines Chinese think tanks as stable and autonomous research organisations that possess independent legal statues, rather than as a part of any government agency or of the CPC.

(15)

the Chinese scholars (except for Zhu, 2009; Deng, 2011). Most Chinese scholars study the classification of American think tanks (e.g.: Wang & Cai, 2001). Deng (2011) categorizes Chinese think tanks into four types, namely the official think tanks, semi-official think tanks, civil think tanks and university-affiliated think tanks. Deng (2011) defines official think tanks as research institutes that are set legally or traditionally within the government to provide policy makers with policy advice. Semi-official think tanks are not officially affiliated with the government; they are however in fact a part of government agencies (Deng, 2011). Deng’s (2011) classification is perceived to be unsuitable for this study for two reasons. Firstly, she regards government- and the CPC-affiliated policy institutes as think tanks. This contradicts the definition of think tanks in this study. Secondly, the boundaries between different types of think tanks are blurry. For example, ‘traditionally within the government’ and ‘in fact a part of government agencies’ are fuzzy descriptions. It is difficult to identify and map the Chinese think tanks using this taxonomy.

Zhu (2009) divides Chinese think tanks into two types by using formal and legal organizational identities. According to the General Principles of the Civil Law of the People’s Republic of China, there are five types organizational identities in China, namely the jiguan (government agencies), shiyedanwei (public institutions), minban fei qiye danwei (civil nonprofit institutions, CNIs), shehuituanti (social groups), and qiye (enterprises) (Zhu, 2009). Zhu (2009) endorses that think tanks are legally independent of the government; think tanks that are registered as government agencies are therefore excluded. Based on legal organizational identities, Zhu (2009) breaks Chinese think tanks into two types: “public institution (semi-official) think tanks” and “civil think tanks” (p. 338). Semi-official think tanks are research organisations that are registered as public institutions; civil think tanks include research organisations that are either CNIs, enterprises (profit organizations that do policy analytic and advisory activities), or university-run policy institutes. The use of legal organizational identities not only considers the Chinese context, but it is also operational and practical.

(16)

Author(s) Categories Assessment

Weaver (1989)

1. universities without students: long-term academic research; book-length research outcomes

1. Book-length research products are neither practical for policy makers nor understandable to the public.

2. Think tanks in reality may undertake multiple

characteristics simultaneously. 2. contract research organisations:

contract with agencies; research reports targeting at government agencies 3. advocacy tanks: a strong focus of

policy or ideology; synthesize existing research

McGann (2000; 2016)

independent think tanks (intellectually,

financially, and legally): 1. academic-diversified and

academic-specialized think tanks 2. contract research organisations 3. advocacy think tanks

4. policy enterprise organisations

1. The first type is not applicable to Chinese think tanks

because of their lack of independence of the government and the CPC. 2. The second type contradicts

the definition of think tanks of this study. This study defines Chinese think tanks as research institutions with independent legal statues; they are not affiliated to any other organizations.

affiliated think tanks:

1. party-affiliated think tanks

2. government sponsored think tanks 3. private think tanks (for profit) 4. university-based think tanks

Deng (2011)

1. official think tanks: legally or traditionally within the government

1. Boundaries between different types are blurry.

2. semi-official think tanks: not officially affiliated with the government but in fact a part of government agencies

3. civil think tanks

4. university-affiliated think tanks

Zhu (2009)

1. semi-official think tanks: public institution think tanks

1. The use of formal and legal organizational identities provides a basis of

categorization for this study. 2. CNI think tanks: civil nonprofit

institution think tanks

3. enterprise think tanks: companies that make profit by doing policy research and advisory activities

4. university-run policy institutes

(17)

Zhu’s (2009) classifying method provides this study with a basis of Chinese think tanks’ classification. Since this study defines Chinese think tanks in terms of independent legal status, using legal organizational identities to categorize think tanks is logical. This study thus breaks Chinese think tanks into two types, namely the semi-official think tanks and civil think tanks. Semi-official think tanks are research organisations with the organizational identities of public institutions. They are by law established, partly/fully funded, and supervised by the government and the CPC. Civil think tanks are centered in CNI and enterprise research organisations. CNI think tanks are non-profit organisations that are established and funded by actors including the government, enterprises, social organisations, and private citizens. Enterprise think tanks are research organisations that are operated as companies and make profits by doing analytical and advisory activities. It is noteworthy that this study excludes university-run policy institutes that are affiliated with universities, as they do not possess independent legal statuses. To conclude, this study adopts Zhu’s (2009) classifying method, categorizing Chinese think tanks into two types. One is semi-official think tanks that consist of public institutions, the other is civil think tanks that are composed of CNI think tanks and enterprise think tanks.

Figure 1: Chinese think tanks: classification

Semi-official think tanks are closely related to the government. Since they are public institutions established and sponsored by the state, they accept government funds as one of their main sources of

Chinese policy research organisations governmental policy institutes university-run policy institutes Semi-official

think tanks Civil think tanks

CNI think tanks public institution think tanks enterprise think tanks

(18)

funding. This makes them financially dependent on the government. Additionally, semi-official think tanks are usually supervised by government agencies in their operations. According to the related laws and regulations, supervising agencies are responsible for the review of establishment, modification, and cancellation of registration of the supervised organisations, and for inspecting the implementation of the CPC guidelines and policies, states laws and regulations by supervised organisations. Besides, semi-official think tanks are most frequently headed by government officials. It can be said that the policy research of semi-official think tanks is to some degree shaped by the government. Nevertheless, they enjoy a comparatively high level of autonomy compared to official policy institutes for two reasons. Firstly, they have other funding sources other than the regular government funds (e.g.: from enterprises), which makes them less financially dependent on the government than official policy institutes. Secondly, they can decide their research topic, meaning that they can accept research tasks from the government, other organisations, such as enterprises and NGOs, and even international organisations.

Civil think tanks are less associated with the government compared to semi-official ones. Generally speaking, they have diverse funding sources, such as funds from enterprises and international organisations. This financial independence guarantees civil think tanks a complete autonomy in deciding research topic, conducting research, publishing research results, and choosing personnel without restriction from the government and the CPC. Although some of the CNI think tanks are also ratified and supervised by government agencies as semi-official think tanks, this supervisor-supervisee relationship is suggested to be rather nominal and loose (Zhu, 2009; Zhu, 2011).

As previously introduced, this study focuses on Chinese semi-official think tanks. In view of the criticism that Chinese think tanks are too dependent on the government and political parties to be ‘true’ think tanks, this study investigates the content and extent of advocacy activities of semi-official think tanks. Because Chinese semi-official think tanks are by nature more closely related to the government and the CPC, the result of this study can be a more relevant response to the common critique.

(19)

2.3 The difference between Chinese think tanks: administrative linkages

Chinese think tanks not only differ in their legal organizational identities: public institutions, CNIs, or enterprises, but they also vary in the strength of their “administrative linkages” with policy makers (Zhu, 2009, p. 341). By defining administrative linkages as the administrative rank of supervising units, Zhu (2009) claims that administrative linkages are shaped by the connection between think tanks and policy makers, and this connection is essential for Chinese think tanks’ success in influencing policymaking. Zhu’s (2009) survey of the heads of Chinese think tanks reveals that the strength of administrative linkages differentiates semi-official and civil think tanks. Semi-official think tanks in Zhu’s (2009) sample averagely had stronger administrative linkages with policy makers, while civil think tanks averagely had weaker administrative linkages.

Evolving from Zhu’s argument (2009), this study argues that administrative linkages can be understood as Chinese think tanks’ direct access to policy makers. Administrative linkages can affect the chance of think tanks’ success in influencing policymaking, and the strength of these linkages is determined by the administrative rank of think tanks’ supervising units. To illustrate, both Chinese semi-official think tanks and CNI think tanks have government agencies as their individual supervising units. This supervisor-supervisee relationship constitutes think tanks’ direct access to policy makers, meaning that these think tanks can directly access to government officials in their daily operation. That being said, both semi-official think tanks and civil think tanks (except enterprises think tanks) have administrative linkages. However, administrative linkages can vary in strength. The strength of administrative linkages is determined by the administrative rank of the supervising units. Since government agencies play the role of supervising units, their administrative rank (e.g.: state, ministry, bureau, division, etc.) directly determines the level of government officials think tanks can access to. The higher level of the administrative rank is, the higher level of government officials that think tanks can access to, the stronger the administrative linkages of think tanks are perceived to be. And the stronger the administrative linkages are, the more likely think tanks are to be heard and their inputs to be adopted by policy makers.

(20)

Figure 2: Chinese think tanks’ administrative linkages with policy makers

Zhu (2009) proposes and empirically tests a relationship between Chinese think tanks’ types and their administrative linkages with policy makers. By interviewing Chinese think tanks’ heads about their administrative ranks, Zhu (2009) concludes that Chinese semi-official think tanks generally have stronger administrative linkages with policy makers than civil think tanks. Zhu’s (2009) study however has a limitation. He defines administrative linkages as the administrative rank of supervising units, yet he measures administrative linkages by looking at the administrative rank of think tanks’ heads instead of their supervising units’. Despite the disadvantage, this study builds on Zhu’s (2009) study and argues that Chinese think tanks’ types are correlated with their administrative linkages with policy makers, and semi-official think tanks generally have stronger administrative linkages than civil think tanks. The reason for this argument is that semi-official think tanks are all established and sponsored by the government, while civil think tanks can be ratified by diverse actors including government agencies, public institutions and enterprises. In this manner, it can be argued that semi-official think tanks are more closely related to the government than civil think tanks, and they therefore have stronger administrative linkages with policy makers.

As this study focuses on Chinese semi-official think tanks, it is noteworthy that this study expects that there is a variation in the strength of administrative linkages within Chinese semi-official think tanks. Semi-official think tanks are individually supervised by different government agencies with varying levels of administrative rank. For instance, Chinese Academy of Engineering is a renowned public institution think tank in China. It is directly supervised by the Chinese State Council, which has an administrative level of state. Another famous Chinese semi-official think tank, Nanjing Academy of

government agencies Chinese think tanks supervisor-supervisee relationship administrative linkages

(21)

Social Sciences, is supervised by the Nanjing city government, which has a lower administrative level of bureau. Chinese Academy of Engineering therefore has a stronger administrative linkage with policy makers than Nanjing Academy of Social Sciences.

2.4 Chinese think tanks: roles and activities

To understand the increasing significance of think tanks, one needs to look at think tanks’ role in policymaking process and the corresponding activities they are involved in to fulfil their role. As ‘role’ and ‘function’ are used interchangeably by scholars (e.g.: Weaver, 1989; McGann, 2000; Stone, 2005), this study uses the term of ‘role’. Weaver (1989) explores five common roles that American think tanks engage in. Firstly, think tanks provide a “source of policy ideas” for policy makers (Weaver, 1989, 568). In addition to conducting policy research, think tanks need to prommote their policy ideas to policy makers and the public. Some of their policy ideas may not be preferred by policy makers out of political reasons, hence, think tanks need to disseminate their policy ideas such that these policy ideas can be eventually accepted and enacted by policy makers. Secondly, think tanks are the “source[s] of evaluator of policy proposals” (Weaver, 1989, p. 568). What American think tanks frequently do is to advance policy ideas by evaluating a specific policy proposal, rather than develop a new approach to a policy topic. Thirdly, think tanks can function as “evaluator of government programs” (Weaver, 1989, p. 569). They can do research to evaluate an ongoing government program, either at their own initiatives or as required by the government. This role helps to oversee and correct government programs. Fourthly, think tanks supply personnel and expertise to the government. Because of a highly frequent turnover of top-level officials in the government, the government has a great demand for personnel with substantive expertise. Think tanks can help to meet this demand by cooperating with the government or providing the government with the desired personnel. Lastly, think tanks play the role of “authoritative source of information and opinion” for the public (Weaver, 1989, p, 569). They are often called upon (usually by news agencies) to comment on a specific policy topic, and their answers are regarded as “punditry” (Weaver, 1989, p. 569). These five roles by Weaver (1989) are helpful for understanding think tanks’ significance in a general way. However, how to recognize a role that think tanks are playing remains unclear, as Weaver (1989) does not elaborate on the activities that think tanks participate in to undertake their roles.

(22)

McGann (2000) touches upon the activities think tanks are involved in to fulfil their roles. According to him, think tanks usually play roles, such as “serving as an informed and independent voice in policy debates, transforming ideas and emerging problems into policy issues, interpreting issues, events, and policies for the electronic and print media, thus facilitating public understanding of domestic and international policy issues”, and so on (McGann, 2000, p. 5). Achieving these roles involves activities of “framing policy issues, researching and writing books, articles, policy briefs, and monographs, media appearances, and speeches”, etc. (McGann, 2000, p. 6). McGann’s (2000) theory of think tanks’ roles and activities is more specific compared to that of Weaver (1989). Yet, McGann (2000) does not clarify what activities constitute a certain kind of role. That is to say, one can observe the activities that think tanks are involved in but still cannot discern the role that think tanks are taking.

Zhu (2011) develops an analytical framework of Chinese think tanks’ roles and the corresponding activities. Zhu (2011, p. 672) claims that Chinese thing tanks generally take up three roles in public policy process, namely the advisors to policy makers, academics in universities and policy institutes, and advocates in the public domain; and these roles can be discerned by evaluating think tanks’ activities. To influence the policy process directly, Chinese think tanks engage in advisory activities, such as applying for research projects that are initiated and sponsored by the government, submitting policy reports to policy makers either through an official or private way, participating in consultation meetings that are held by the government, and giving lectures to policy makers (Zhu, 2011). In addition to advisors to policy makers, Chinese think tanks are also active in public sphere as advocates. The Chinese government has established a filter mechanism, which provides think tanks with an opportunity to draw the attention of policy makers. To elaborate, this filter mechanism is a 24-7 surveillance system that automatically monitors information from mass media, internet and publications. Information that is perceived to be progressive or sensitive is automatically extracted and collected by this system. This system then reports the documented progressive or/and sensitive information with its authors to the government. The government will decide how to deal with this information and its authors depending on the content of information. Simply put, it can be seen as a regulatory mechanism of speeches. Although this filter mechanism is commonly criticised by scholars

(23)

as a strict control of intellectual life from the government and the CPC, it can be regarded as a channel of approaching policy makers for think tanks. Think tanks, especially those with progressive positions can take advantage of media and internet to get their policy ideas noticed and heard by policy makers. Therefore, Chinese think tanks that play the role of advocates are involved in advocacy activities of accepting interviews, publishing press reports, establishing website, publishing ideas online, and delivering lectures to the public (Zhu, 2011). Lastly, Chinese think tanks can take up an academic role, which this study regards as a basic function of all think tanks. This role entails activities ranging from publishing policy reports or papers in journals, to attending and holding academic forums for the exchange of policy ideas. Zhu’s (2011) framework establishes a clear connection between think tanks’ roles and activities, which is beneficial to identifying the role think tanks play in policy process and further to understanding the significance of think tanks.

Functions Academics Advisors Advocates

Corresponding Activities 1. Writing and publishing academic papers 2. Publishing policy

reports and books 3. Holding and

participating in academic forums 4. Teaching and

training

1. Applying for the government research projects 2. Submitting policy reports to policy makers 3. Participating in the government consultation meetings 4. Giving lectures to policy makers 1. Accepting interviews by mass media 2. Publishing press reports 3. Establishing website 4. Delivering lectures to the public 5. Blogging on the internet

Table 3: Chinese think tanks’ roles and activities (Zhu, 2011)

As this study explicitly focuses on Chinese think tanks’ advocacy role and the related activities, Zhu’s (2011) framework is adopted. Zhu (2011) establishes a clear connection between think tanks’ roles and activities, which makes it possible to discern the role that think tanks are playing by investigating activities that they engage in.

(24)

2.5 Possible determinants of the extent of advocacy activities of Chinese think tanks

The insider/outsider theory gives insights into the determinants of the extent of think tanks’ activities. The term of insider/outsider was originally developed to describe the differentiated access to policy makers by interest groups in Britain (Maloney, Jordan & McLaughlin, 1994). Insider groups were defined as groups that had the direct/privileged access to policy makers, in the sense that they could approach policy makers directly and they knew “where to go for information on their initiative and knew how to get it” (Schattschneider, 1935, p. 166, cited in Maloney et al., 1994). On the contrary, outsider groups were not granted this direct/privileged access, they thus knew nothing at all until it was too late (Maloney et al., 1994). The insider/outsider distinction can be applied to Chinese think tanks. Chinese think tanks vary in the strength of their administrative linkages with policy makers, which can be equated with the differentiated direct access in insider/outsider theory. This study argues that Chinese think tanks with stronger administrative linkages can be regarded as insider groups in policy process, while think tanks with weaker administrative linkages can be seen as outsider groups in policy process.

Literature on insider/outsider theory has supported that the direct access affects interest groups’ preference for strategies of influencing policy process. Grant (2004) combined the distinction of insider/outsider groups with group strategies in 1978. Defining strategies as the combination of diverse modes of action that interest groups undertake to achieve their goals, Grant (2004) claims that the distinction of insider/outsider is essentially a differentiation of strategies that interest groups choose (Maloney et al., 1994). Insider groups preferably utilise insider strategies of participating in consultation process with policy makers; they seek to directly influence policymaking by contacting policy makers (Binderkrantz, 2005). Oppositely, outsider groups without privileged access indirectly influence policymaking by involving media and the public. They predominantly resort to outsider strategies of using mass media and mobilising the public (Binderkrantz, 2005 p. 696).

There are different components of insider and outsider strategies. Insider strategies consist of “administrative strategy” and “parliamentary strategy” (Binderkrantz, 2005, p. 696). Administrative

(25)

public officials (e.g.: “relevant minister”, “public servants”), utilising “public committees”, “responding to requests for comments” (Binderkrantz, 2005, p. 696). Parliamentary strategies target at political parties and politicians. Actions include contacting “parliamentary committees”, “party spokespersons”, parliament members, and “party organisations” (Binderkrantz, 2005, p. 696). Outsider strategies are divided into “media strategy” and “mobilization strategy” (Binderkrantz, 2005, p. 696). A media strategy refers to actions of cooperating with mass media. For example, outsider groups can choose to contact or write letters to the press, publish press reports or research reports, and hold press conferences (Binderkrantz, 2005, p. 696). A mobilization strategy involves actions, such as holding “public meetings” and “conferences”, and “arranging strikes” or “public demonstrations” (Binderkrantz, 2005, p. 696). Grant (2004) further implies that insider groups that employ insider strategies are more likely to be successful in getting their input to be adopted by policy makers compared to outsider groups that have to utilise outsider strategies. Indeed, as Binderkrantz (2005) states, most scholars consider outsider strategies to be a fall back option.

The division of insider/outsider strategies coincides with the different types of Chinese think tanks’ activities. As mentioned before, think tanks can be considered as a type of interest groups as they have the goal of influencing policymaking. Think tanks engage in diverse activities to achieve their goal. For example, Chinese think tanks engage in advisory activities that can influence policy makers in a direct way (e.g.: contacting policy makers, participating in government consultation meetings), which corresponds to an insider strategy. Similarly, advocacy activities that focus on mobilizing the pubic by the means of mass media and internet fall into the category of media strategy in insider/outsider strategies.

As discussed above, insider groups are expected to use more insider strategies while outsider group resort more to outsider strategies (Grant, 2004; Binderkrantz 2005). In the same manner, it can be argued that Chinese think tanks with stronger administrative linkages, which equate to insider groups with the direct access, prefer more advisory activities of directly influencing policy makers. They therefore have a smaller extent of advocacy activities and function more as advisors. Chinese think tanks with weaker administrative linkages, that resemble outsider groups without privileged access, choose to engage in more advocacy activities of indirectly influencing policy makers. They act

(26)

more as advocates and have a larger extent of advocacy activities. The role of academics in policy institutes is treated as a basic function for all Chinese think tanks, regardless of their administrative linkages. The reason for this has been stated in the first part of this section: think tanks’ principal function is to conduct policy analytic practices. As such, the following hypotheses are formulated:

Hypothesis 1a: Chinese think tanks’ administrative linkages determine the extent of their advocacy activities.

Hypothesis 1b: Chinese think tanks with weaker administrative linkages engage in more advocacy activities than ones with stronger administrative linkages.

Nevertheless, there is literature suggesting that the direct access does not affect interest groups’ preference for strategies of influencing policy process. As Binderkrantz (2005) observes, most interest groups combine insider strategies of directly approaching policy makers with outsider strategies of mobilizing mass media and the public. In fact, insider groups are increasingly involved in activities of media campaigns and member mobilizations, as outsider strategies are perceived to be more effective compared to insider strategies. Likewise, Popović’s (2017) study of Chinese citizens’ groups reveals that advocacy groups in China use a combination of advisory activities and advocacy activities. To influence policy process, Chinese advocacy groups simultaneously engage in activities of contacting policy makers and activities of increasing their public support, no matter whether they have privileged access to policy makers. This means, the direct access to policy makers does not determine whether Chinese advocacy groups prefer advisory activities or advocacy activities. Therefore, an alternative hypothesis is developed as the following:

Hypothesis 2: Chinese think tanks’ administrative linkages do not determine the extent of their advocacy activities.

Another potential determinant of the extent of Chinese think tanks’ advocacy activities is regional knowledge capacity. Zhu (2011) claims that Chinese think tanks’ advocacy activities can be

(27)

determined by the regional knowledge capacity, which is defined as the ability of potential audience to procure, digest, and communicate knowledge. Think tanks located in different cities are faced with audience with diverse knowledge capacity, which determines the role that think tanks predominantly play. In regions where there is a high level of knowledge capacity, residents are more willing to embrace, obtain, discuss and absorb new ideas than residents in regions with a low level of knowledge capacity. Hence, Chinese think tanks that are operated in high knowledge capacity regions encounter less resistance when publishing and communicating their new research results or policy ideas, they are consequently more involved in advocacy activities to mobilise the public. In contrast, think tanks located in regions with a low knowledge capacity are confronted by more resistance when they try to communicate a new policy idea to the public. They therefore engage in less advocacy activities. Hu and Xiong (1998, cited in Zhu, 2011) divide Chinese regional knowledge capacity into four categories (see Table 4). Based on this, the following hypotheses are developed:

Hypothesis 3a: Regional knowledge capacity determines the extent of advocacy activities of Chinese think tanks.

Hypothesis 3b: Chinese think tanks located in high knowledge capacity regions engage in more advocacy activities than ones located in low knowledge capacity regions.

regional knowledge capacity Cities in China

high Beijing, Shanghai, Tianjin, Guangdong, Fujian, Jiangsu upper-middle Hainan, Liaoning, Zhejiang

lower-middle Shandong, Hebei, Jilin, Hubei, Heilongjiang, Shanxi (in west China)

bottom Hunan, Shanxi (in central China), Henan, Xinjiang, Chongqing, Inner Mongolia, Gansu, Ningxia, Sichuan, Guangxi, Yunnan, Qinghai, Guizhou, Tibet, Jiangxi, Anhui

(28)

Radaelli’s (1999) study of expertise in politics provides a different angle: the policy areas that think tanks are active in. Radaelli (1999) argues that the adoption of expertise in politics is decided by two dimensions, namely the salience, and the uncertainty of policy issues. As regards the dimension of salience, policy issues can be divided into two types. One is “opaque” policy domain, where policy issues are confined to policy makers and not visible to the public (Radaelli, 1999, p. 763). The other is salient policy domain, where media and the public are allowed to discuss and influence policymaking. Uncertainty refers to the complexity of policy issues. For uncertain and complex policy issues, they demand a great deal of expertise to provide insights and to solve them. In contrast, policy issues that are in low level of uncertainty are less complex. They therefore have a low demand for knowledge and experts. Taken together, when policy issues are nether visible (a low level of salience) nor complex (a low level of uncertainty), there will be “bureaucratic politics”–a political competition of power–rather than a learning process (Radaelli, 1999, p. 764). Under such circumstances, media and the public do not have a say over the policy issues; knowledge and expertise are not favoured by politicians. On the contrary, for think tanks that are active in salient and uncertain policy domains, they can play the roles of both advisors and advocates. They can influence policymaking either by providing expertise for policy makers or mobilizing the public. As for think tanks dealing with opaque and uncertain policy issues, they are logically act more as advisors to policy makers, as they are needed to offer scientific knowledge while these policy issues per se are not open to public opinion. For think tanks engaging in salient and certain policy issues are more likely to engage in advocacy activities to appeal to public supports for their policy ideas. As such, the last set of hypotheses is formulated:

Hypothesis 4a: The type of policy areas Chinese think tanks are active in determine the extent of their advocacy activities.

Hypothesis 4b: Chinese think tanks that are active in salient policy types engage in more advocacy activities than ones active in non-salient policy types.

(29)

framework to give insights into the determinants of the extent of advocacy activities of Chinese think tanks.

Figure 3: Theoretical framework of explaining determinants of the extent of advocacy activities of Chinese think tanks

This study has developed a series of hypotheses. They are summarized as the following:

Hypothesis 1a: Chinese think tanks’ administrative linkages determine the extent of their advocacy activities.

Hypothesis 1b: Chinese think tanks with weaker administrative linkages engage in more advocacy activities than ones with stronger administrative linkages.

Hypothesis 2: Chinese think tanks’ administrative linkages do not determine the extent of their advocacy activities.

Hypothesis 3a: Regional knowledge capacity determines the extent of advocacy activities of Chinese think tanks.

Hypothesis 3b: Chinese think tanks located in high knowledge capacity regions engage in more advocacy activities than ones located in low knowledge capacity regions.

Hypothesis 4a: The type of policy areas Chinese think tanks are active in determine the extent of their advocacy activities.

Hypothesis 4b: Chinese think tanks that are active in salient policy types engage in more advocacy Administrative linkages

Regional knowledge capacity

Type of policy areas

The extent of advocacy activities

(30)

Determinants of Chinese Think Tanks’ Advocacy Role activities than ones active in non-salient policy types.

3. Research Design and Data Collection

This section describes how the proposed theoretical framework in section 2 has led to the corresponding research design. The first part of this section details how the relevant concepts derived from the theoretical framework are operationalized and measured. The next part explains how the research design fits with the research question of this study. The subsequent part introduces the source of data and gives details of data collection.

3.1 Operationalization and measurement of key variables

3.1.1 Independent variables

The first explanatory variable – independent variable of interest – of this study is think tanks’ administrative linkages with policy makers. As discussed in the previous section, administrative linkages refer to think tanks’ direct access to policy makers, which can be indicated by the administrative rank of think tanks’ supervising units in the Chinese context. Therefore, this study operationalizes administrative linkages as the administrative rank of think tanks’ supervising units. According to Temporary Regulations for State Civil Servants of the People’s Republic of China, there are twelve administrative ranks in China. They are the national and sub-national level, provincial-ministerial and sub-provincial/ministerial level, bureau-director and deputy-bureau-director level, division-head and deputy-division-head level, section-head and deputy-section-head level, section member level, and ordinary staff level. In this study, Chinese administrative ranks are merged into five levels and coded as the following: 5 denotes national level; 4 denotes ministerial level; 3 denotes bureau level; 2 denotes division level; 1 denotes section level or below or none government supervising units.

Another explanatory variable is regional knowledge capacity. Chinese think tanks are located in diverse regions with varying levels of regional knowledge capacity. According to Hu and Xiong (1998, cited in Zhu, 2011), Chinese regions can be divided into four groups in terms of their regional knowledge capacity (see Table 4). This study thus operationalizes and measures regional knowledge