Policy Coalition Building in an Authoritarian Legislature: Evidence From China’s National Assemblies (1983-2007)

Abstract

Recent studies of authoritarian legislatures underscore the importance of institutions for cooptation and information collection, but many still consider authoritarian legislatures rubber stamps in policy making. We argue that authoritarian legislatures could be important arenas of the contestation reflected in delegates’ bills and proposals instead of their voting outcomes. Specifically, government agencies use authoritarian legislatures to build policy coalitions to advance their policy agendas. Delegates serve as proxy fighters for key party and government elites with different policy preferences. We provide evidence based on an original dataset containing education-related bills and policy proposals submitted to both Chinese national assemblies between 1983 and 2007. We identify the existence of the policy coalitions, and find coalition building is more intensified prior to the promulgation of education-related laws. We then employ network analysis to illustrate the channels behind policy coalition network.

Keywords

China legislative studies nondemocratic regimes

Similar to many legislative institutions in authoritarian regimes, China’s national assemblies—the National People’s Congress (NPC) and Chinese People’s Political Consultative Conference (CPPCC)—are often viewed as ceremonial institutions. Although some scholars have suggested that Chinese legislatures have been more assertive in congressional oversight in the last two decades,¹ Melanie Manion (2008) pointed out the “most lively congresses are found not at the center of power in Beijing but in the localities” (p. 608). A close look at the submissions of NPC bills and CPPCC policy proposals reveals a surprisingly vibrant dynamic, however, suggesting the national assemblies are anything but ceremonial. According to Tanner (1995), bills submitted to the NPC should have already received approval from the top leadership after negotiations behind closed door. Despite over 16,000 NPC bills² introduced in the NPC between 1983 and 2007, only 486 bills have been promulgated. Meanwhile, CPPCC delegates submitted nearly 70,000 policy proposals over the course of 25 years. The dynamics of bill and proposal submissions present an interesting puzzle: Why do NPC and CPPCC delegates actively submit bills and proposals if the national assemblies are only ceremonial?

Recent studies have focused mostly on the functional view of authoritarian legislatures, which mainly concerns the ways through which these institutions facilitate authoritarian rule. These studies primarily maintain two mechanisms: cooptation and information collection. Specifically, some argue that formal institutions facilitate cooptation and power sharing among allies and the opposition;³ others contend that election outcomes in authoritarian regimes reveal critical information on mass preferences, enabling ruling elites to distribute public spending and spoils to targeted populations to assure regime survival.⁴

These mechanisms, however, cannot adequately explain delegate behavior in the NPC and CPPCC. In particular, the Chinese regime has developed much better tools for cooptation and information gathering than what the NPC and CPPCC may offer. In terms of cooptation, Chinese ruling elites instead of national assemblies have provided their allies with preferential treatment in the realms of business as well as government and party offices. The emergence of the “princeling” class is a prominent example—various reports have shown that its members have accumulated massive wealth and political power.⁵ The depth and breadth of economic and political networks among the relatives of the members of Central Committee and Politburo in publicly traded firms offer another telling example.⁶

When it comes to gathering information on mass preferences, the Chinese government has built more efficient channels to assess local conditions and discontent than the information that the NPC and CPPCC may offer. For example, Manion (2015) argues local People’s Congresses help broker local knowledge into the provision of local public goods. Dimitrov (2014) maintains that the Chinese government, similar to other Communist autocracies, has used internal government assessments of its governance through citizens’ petitions to identify and address public discontent. Meanwhile, public protests could serve as another channel for the Chinese government to gather information concerning local stability (Huang, Boranbay, & Huang, 2019; Lorentzen, 2013, 2014). Finally, the rise of information technology has prompted the Chinese government to build various national and local apparatuses to manage and collect information over the Internet (King, Pan, & Roberts, 2014).

Although cooptation and information might have been the original functions when the NPC and CPPCC were created in China, these mechanisms do not seem to play a major role in these two national assemblies today. In this article, we deviate from the functional view by shifting to a new analytical lens through which we view the national assemblies as one of the battle grounds for policy contestation among regime insiders. We contend that bureaucratic influence plays an important role in the dynamics of legislative activities. The logic of our argument is based on the premises that policy making is often conflictual, and government bureaucracies have inherently different policy preferences. Hence, persistent structural conflicts exist among bureaucracies and other functioning groups. These structural conflicts are not unique to the Chinese system: previous scholars who studied Soviet politics have documented structural conflicts among the existing party and government systems that play a significant role during the policy making process (Hough & Fainsod, 1979; Skilling & Griffiths, 1971; Skilling, 1983).

Because of these structural conflicts, we argue that authoritarian legislatures provide an opportunity to allow regime insiders to build policy coalitions, which help advance their policy agendas. Building policy coalitions in the national assembly serves two important objectives. First, the success of policy coalitions in the national assembly could lead to concrete laws and regulations, which address the credible commitment problem for any policy bargains made in closed-door meetings. Second, policy coalitions in the national assembly exert pressure on the oppositions in the public domain and draw attention from ruling elites.

To evaluate our claims, we offer evidence from an original dataset containing all the NPC bills and CPPCC policy proposals concerning compulsory education between 1983 and 2007. The empirical results of our article are threefold. First, we demonstrate the existence of policy coalitions in both Chinese national assemblies through coordinated efforts in bill and proposal submission; these coalitions intensify prior to the promulgation of education-related laws. Second, delegates with employment ties to central government agencies, particularly ties to the Ministry of Education (MOE), are more likely to participate in policy coalitions concerning education; however, the more intensified policy coalition building also include delegates who have no ties to the MOE. Third, we use network analysis to identify two important channels—party affiliation and geographic locations—through which delegates build policy coalitions.

Our article makes two main contributions. First, this article is among the emerging scholarship on authoritarian legislatures that relies on micro-level evidence to uncover important legislative dynamics (e.g., Desposato, 2001; Malesky & Schuler, 2010; Noble, 2017). We advance the study of formal institutions under authoritarianism by surpassing the functional view. By investigating legislative activities and delegate behaviors, our study casts new light on the study of authoritarian legislature. We show that an authoritarian legislature could become one of the venues that allow regime insiders to set policy agendas and win policy battles. Delegates in the national assembly may serve as proxy fighters on behalf of various government agencies. In democracies, legislative sessions are often considered the platform for political elites’ strategic interaction to advance their policy preferences. By the same token, legislative behaviors in authoritarian regimes could reflect policy struggles among government agencies. Therefore, our article suggests that future micro-level study of legislative behaviors may reveal new insights about elite politics in authoritarian regimes.

Second, our article extends previous studies on interest groups in Communist regimes as well as elite policy making in China. In particular, we provide a unified theoretical framework that connects legislative behavior and bureaucratic interests in authoritarian regimes. Previous scholars of Soviet politics have proposed an interest group model (Skilling, 1966), arguing that the policy making process is deeply influenced by differential institutional interests during the post-Stalin era. Meanwhile, several researchers have laid the groundwork for theorizing the process of elite decision making in China (e.g., Lieberthal & Lampton, 1992; Shih, 2008; Shirk 1993). One prominent theory is the argument of “fragmented authoritarianism” that describes policy making as involving multiple bureaucratic interests in China (Lieberthal & Oksenberg, 1988). In this context, scholars have argued that Chinese legislative institutions such as the NPC play a limited role in policy making (O’Brien, 1990, 1994; Tanner, 1995, 1999); however, recent studies have suggested changing legislative behavior at the local level (e.g., Hou, 2015; Manion, 2014) as well as changing the nature of the policy making process in China (Mertha, 2009). We show that changes in legislative behaviors also occurred at the national level. We argue that “fragmented authoritarianism” and interest group influence from regime insiders manifest themselves in the Chinese national assemblies, which have been increasingly used by various central government agencies to signal their preferences and build policy coalitions in recent years.

The roadmap of this article is as follows: the next section offers a theoretical framework of legislative behavior and bureaucratic influence in authoritarian national assemblies. We then describe our research design and data collection, including a variety of empirical evidence. We conclude the article by offering some implications.

A Theoretical Framework

In this section, we focus mainly on the motivations and formation of policy coalition building in the national assemblies in authoritarian regimes. We first discuss the logic of bureaucratic influence through policy coalition building in authoritarian national assemblies. We then detail delegates’ incentive structure in participating in the policy coalition building.

The Logic of Bureaucratic Influence

The premise of our theoretical framework hinges on policy struggles among regime insiders. Policy struggle is particularly intense for a class of authoritarian regimes that have created a massive bureaucratic system. In Communist regimes, for instance, Hough and Fainsod (1979) point out that the strongest political actors below the leadership level are often vertical branches, not always regional officials. Vertical conflicts among different government agencies are in part the result of the central planning economic system in Communist regimes. The complexity of the economy made it clear to the Soviet leadership that they must allow a reverse stream of influence from its massive bureaucracy. Hence, group interests are promoted, and conflicts often occur during the policy making process. In a similar vein, Lieberthal and Oksenberg (1988) observe deeply divided bureaucratic interests during the policy making process in China, and they coined the term fragmented authoritarianism to describe bureaucratic politics in China’s policy making. Although the Chinese government has experienced multiple organizational reforms aiming to create an administrative state since the 1980s, structural conflicts among government bureaucracies persist because of the vertical structure of many government bureaucracies (Yang, 2004).

Given the existence of structural conflicts among government agencies, bureaucrats must resolve their differences in policy making. The conventional wisdom is that an authoritarian legislature is inconsequential in policy making. Key policies and legislations are driven by internal debates in closed-door meetings, and disagreements would have been resolved among ruling elites and government agencies before they reach the authoritarian legislature.

We certainly agree authoritarian legislatures often lack the open debates and contestation that are commonly observed in democratic legislatures. Nonetheless, we argue that an authoritarian legislature could become an important battle ground for policy debates among government agencies for two reasons. First, the closed-door policy bargains result in credible commitment problems in authoritarian politics (Myerson, 2008; Svolik, 2012). By transforming policy preferences into formal laws and regulations through authoritarian legislatures, government bureaucracies could mitigate the credible commitment problem. Second, those government agencies that hold a disadvantageous position during policy debates in closed-door meetings could seek alternative channels, such as the national assemblies, to generate pressure on their opposition to engage in policy making by drawing the attention of the ruling elites.

More importantly, we contend that effective bureaucratic influence requires government agencies to build a policy coalition in the national assembly. We define policy coalition as a temporary coordinated effort among political actors inside and outside the government agency i to advocate a particular policy position that benefits only government agency i in the national assembly.⁷ It is not surprising that delegates with ties to a particular government agency typically put forward proposals that fall within this government agency’s domain. Nonetheless, a more effective policy coalition requires delegates without ties to this government agency to work along with delegates without these ties to submit similar bills and proposals concerning a particular issue in the national assembly. Notably, policy coalitions may not manifest through up-and-down votes on bills and proposals largely because an authoritarian legislature is often highly controlled and bills reach the final voting stage without much debate. We argue, however, that bill and policy proposal submissions become an alternative venue for bureaucratic influence to build policy coalitions. Essentially, these bill and proposal submissions allow government agencies to advocate their policy agendas by gathering support from other regime insiders.

The policy coalition building in the national assembly could be effective in advancing policy making mainly for two reasons. First, ruling elites often have imperfect knowledge of the policy preferences held by various stakeholders (i.e., regime insiders). Bill and policy proposals in the national assembly allow government agencies to publicly express their policy preferences to all the top ruling elites instead of only those who attend the closed-door meetings.

Second, ruling elites also have imperfect information on the support behind certain policy proposals by regime insiders. Building policy coalitions through bills and proposal submissions in an authoritarian national assembly publicly demonstrates the strength of policy support from regime insiders. The revelation on the strength of policy support is credible largely because building policy coalitions is not without risk—ruling elites often discourage or even prohibit building coalition among regime insiders. Hence, government agencies and delegates bear some risk when building and participating in policy coalitions. If a large policy coalition is formed in the national assembly, both the opposition and ruling elites must take notice because the formation of that coalition indicates certain polices are endorsed by a large group of regime insiders beyond bureaucratic boundaries.

The Motivations of Delegates in Authoritarian Legislatures

Although government agencies are inclined to exert bureaucratic influence on the national assembly, the natural question then becomes why do delegates participate in policy coalition orchestrated by a government agency after all? We start with the same assumption in existing studies that delegates in authoritarian legislatures are seeking office for a range of benefits.⁸ For example, studies have shown that serving in the legislature is associated with legislators’ economic gains in both democracies (Eggers & Hainmueller, 2009) and autocracies (Hou, 2015; Truex, 2014); furthermore, many countries have laws assuring some level of immunity for the legislator.⁹ More importantly, serving in the legislature provides a unique opportunity for individuals to access the inner circle of political power, where they can participate in key policy debates and network with other important political actors.

If reelection is a primary goal, appealing to constituents is critical to delegates’ political survival. The constituents of the delegates in an authoritarian legislature are not necessarily the ordinary citizens they supposedly represent; instead, the selectorate of the delegates in authoritarian regimes often comprises ruling elites and parties because the selection process favors elite preferences over mass preferences. Even if legislators are elected locally, signaling their loyalty to their selectorates remains critical because placement on the ballot, not to mention winning local elections, often requires strong endorsement from the ruling party.¹⁰ Hence, many consider an authoritarian legislature a rubber stamp because delegates represent primarily the interests of their de facto constituents.

Although many authoritarian legislatures are likely to act as rubber stamps, we argue that the selectorate of delegates includes a diverse group of elites, ranging from various government agencies to party organizations, who may not always share the same policy preferences (Gallagher & Hanson, 2015). Even if the regime is dominated by one party, the sources of delegates’ political support may originate from disparate factions within the government and party. Hence, delegate behavior is likely to be heterogeneous because delegates respond to different groups of elites, especially in an authoritarian regime with embedded structural conflicts among the various functioning groups. This is particularly the case if the delegates have concurrent employment in government agencies; delegates’ legislative behavior is more subject to the impact of groups of elites, who have the power to influence delegates’ careers.

One strategy by which delegates appeal to their selectorate is casting votes as well as submitting bills and policy proposals on their behalf. Casting the nay vote in an authoritarian legislature, however, is often not a viable option and is highly costly to the delegates because the bills reaching the voting stage have already obtained endorsement from the ruling elites. Alternatively, submitting bills and policy proposals is sometimes less risky because ruling elites may not have reached consensus on certain policy issues. We can thus conceptualize the bills and policy proposals as an alternative “voting process” on policy issues in a highly controlled authoritarian legislature.

These bills and policy proposals allow delegates to advocate their selectorate’s policy agendas in the legislature. In this sense, we argue that delegates in the national assemblies could serve as proxy fighters for key government agencies. This observation is consistent with other studies of authoritarian legislatures. For example, Remington (2001) finds that government ministries were the principal sponsors of legislation in the new Russian parliament. Malesky and Schuler (2010) also show that centrally nominated candidates in the Vietnamese national assembly are less likely to be critical of the regime. In fact, Tanner (1995) points out that central-level party and government organizations are responsible for many bills submitted in the NPC in China.

In summary, we argue that authoritarian legislatures provide an important venue that allows government agencies to build policy coalitions to advance their policy agendas. Note that only a small percentage of bills and proposals result in laws and regulations in both democratic and authoritarian legislatures, but unsuccessful bill submission is not necessarily a wasted effort. Submitting unsuccessful bills and policy proposals constitutes intralegislative signaling in an authoritarian regime, not extralegislative signaling as in democratic regimes.¹¹

Our theoretical framework generates three observable implications. First, if bureaucratic influence exists in authoritarian legislatures, we should observe coordinated policy coalitions through bill and policy proposal submissions on a particular issue that represent the interests of a certain government agency; in addition, the policy coalition intensifies before the promulgation of laws and regulations. Second, delegates’ employment ties to corresponding central government agencies provide the key leverage for bureaucratic influence in the national assemblies; however, policy coalitions become more intensified when delegates without ties to a central government agency take part in the coalition building. Third, the coordination of bill and policy submission operates through certain linkages, such as geographic location and party membership, among delegates with or without employment ties to the government agency.

In the next section, we discuss our research design and data collection based on an original dataset derived from the bills and policy proposals in China’s national assemblies between 1983 and 2007. Subsequently, we report the evidence supporting our hypotheses in three parts. We first identify the policy coalitions in China’s national assemblies. We then analyze the background of those delegates who participate in policy coalitions. Finally, we use network analysis to identify key channels—party affiliation and geographic locations—through which the policy coalition is built in one of the national assemblies.

Research Design

The primary challenges to test our argument comprise the identification of policy coalition and collection of data on legislative activities in authoritarian regimes. Drawing evidence from cross-national analysis is subject to omitted variable bias because legislative behavior is shaped by the institutional design of authoritarian legislatures, which could be the result of many observed and unobserved factors. To address this issue, we explore the within-country variation in the selection of delegates in the national assemblies in China. One unique feature of the Chinese political system is the coexistence of two national assemblies—the NPC and the CPPCC—and they have different selection mechanisms that allow different degrees of bureaucratic influence.

Neither NPC bills nor CPPCC policy proposals have been published systematically because general legal requirements are lacking. We focus on one policy area in our data collection: compulsory education. Our choice is driven by two considerations. First, education is distributive in nature, and a large share of the Chinese government budget is allocated to education.¹² Second, we follow the insights from a study of Soviet politics, in which education policies are argued to be less politically sensitive when compared with other policy areas, such as military spending and social stability; thus, they are more subject to bureaucratic influence (Kelley, 1972).¹³

In this section, we briefly discuss the differences in these two national assemblies. We then introduce our dataset, consisting of NPC bills and CPPCC policy proposals with regard to compulsory education between 1983 and 2007.

Background of NPC and CPPCC

The NPC, founded in 1954, is the highest level of legislature in China. It has the constitutional right to amend the constitution and enact laws as well as to approve annual government budgets and appoint individuals to national political posts. Precisely because China’s constitution stipulates the NPC’s de jure power, the Chinese Communist Party (CCP) maintains tight control over delegate selection and bill submission in the NPC to ensure its monopoly (O’Brien, 1990; Tanner, 1999).

Compared with the NPC, the CPPCC holds weaker political power in China’s political hierarchy largely because it lacks de jure power in law making and political appointments. The CPPCC was initially founded in 1945 as the national legislature during negotiations between the CCP and the Kuomintang. After relegating this function to the NPC in 1954, the CPPCC serves as a “united front” to incorporate elites in various sectors of society, especially from those political organizations other than the CCP, to inform policy making. The CPPCC serves similar functions as a consultative body by providing feedback and suggestions from the delegates to policy makers.

NPC and CPPCC delegates are selected through different processes; thus, the delegates in these two national assemblies are subject to different degrees of bureaucratic influence. NPC delegates are elected by the delegates in the provincial People’s Congress, not directly elected by local residents. Hence, NPC delegates are organized by provinces, and they are supposed to represent the interests of their provinces.¹⁴ In contrast, CPPCC delegates are not elected but selected by the central government and party agencies, and delegates comprise leading figures of sectors and parties in Chinese society in addition to the CCP, such as democratic parties, mass organizations, ethnic groups, and sectors of society. Candidates for the CPPCC are first recommended by their corresponding groups,¹⁵ and then these candidates are evaluated by the CCP’s Organization Department and by the United Front Work Department for CCP and non-CCP delegates, respectively. The CPPCC standing committee approves the list of CPPCC delegates. Notably, serving in the NPC and CPPCC is not a full-time job, and most of these delegates either have concurrent employment elsewhere or they are retired government and party officials. Given the difference in the delegate selection process, CPPCC delegates are more subject to bureaucratic influence.

The submission of an NPC bill (yi’an) and a CPPCC policy proposal (ti’an) also differ. According to NPC regulations, if an individual delegate’s bill aims to move forward to the promulgation process, it must have the cosponsorship of at least 30 NPC delegates.¹⁶ In contrast, CPPCC delegates can submit a policy proposal without any cosponsor. Furthermore, CPPCC delegates enjoy greater freedom in the content of policy proposals that they submit because not all proposals are intended for promulgation into laws.¹⁷ As a result, the members of the CPPCC are more susceptible to bureaucratic influence.

Data on NPC Bills and CPPCC Proposals

We focused on bills and policy proposals concerning compulsory education in both the NPC and CPPCC in our data collection, which took place in two steps. First, we obtained all NPC bills and CPPCC policy proposals concerning compulsory education submitted between 1983 and 2007 through the corresponding committees handling bills and proposals in the NPC and CPPCC, resulting in 209 NPC bills and 1,255 CPPCC policy proposals. Each bill and policy proposal consists of some basic information: the name of the proposal initiator, the title of the proposal, the number of cosponsors, and the year of submission. CPPCC proposals also contain the departments or ministries from which the delegates requested a response to their policy proposals. The NPC bills and CPPCC proposals cover a variety of issues. The NPC bills aim at promulgation resulting in laws; thus, these bills contain issues concerning the entire country instead of a particular region. Meanwhile, the CPPCC policy proposals do not face this restriction; thus, they cover issues both at the national level and at the local level.

Two patterns emerge in Figure 1, which illustrates data on the number of NPC bills and CPPCC proposals concerning compulsory education over the course of 25 years (1983-2007). First, the number of proposals has risen in recent years in both the NPC and CPPCC, especially after 2000. Second, over time, both NPC and CPPCC proposals exhibit cyclical patterns of submission, which broadly correlate in both groups. These patterns suggest that the submissions in both national assemblies are likely to be driven by some coordinated efforts.

Figure 1.

NPC and CPPCC compulsory education proposals (1983-2007).

To the best of our ability, we collected background information on the delegates who initiated these bills and proposals from internal government documents and the Internet. Specifically, we identified the party and province affiliation of the delegates as well as the characteristics of their employment. Note that our data contained only the names of the NPC and CPPCC delegates who initiated the bills and policy proposals, and we had no information on the names of other cosponsors of bills or proposals except the total number of cosponsors. The missing data on cosponsors’ names do not undermine our empirical tests because we aim to identify those individuals with the mobilization capacity to signal their loyalty to their constituents by initiating bills or policy proposals and mobilizing other delegates to cosponsor them.

Empirical Findings: Policy Coalitions in the NPC and CPPCC

In this section, we illustrate the bureaucratic influence through building policy coalitions in the NPC and CPPCC. We report empirical findings based on analysis of NPC bills and CPPCC proposals in two parts. We first demonstrate the existence of policy coalition in China’s national assemblies. We then analyze the background of the delegates who participate in the policy coalitions.

Empirics 1: The Existence of Policy Coalitions

One important aspect of our definition of policy coalition is the temporary coordinated efforts among political actors to advocate a particular policy position that benefits only government agency i in the national assembly. In other contexts, policy coalition can be identified through the number of cosponsors or roll-call votes behind a particular bill. In the Chinese national assemblies, however, these measures are not meaningful indicators of policy coalitions for three reasons. First, casting a nay vote is costly and not very effective in the NPC: When the bills reach the voting stage, they have already received endorsement from the top leadership, and NPC delegates rarely challenge the bills at this stage (Tanner, 1995). Furthermore, the CPPCC delegates do not even have an opportunity to vote on individual policy proposals. Second, all NPC bills must garner 30 or more cosponsors before the standing committee of the NPC can consider proposals for promulgation. As a result, the number of cosponsors for NPC bills is artificially inflated because most of the NPC bills have exactly 30 cosponsors. Third, the names of cosponsors behind a NPC bill or a CPPCC policy proposal do not constitute public information even among the delegates; thus, cosponsors receive less credit from their selectorates. In contrast, the identity of the lead sponsor who submits a bill or proposal is always public information; hence, CPPCC delegates often prefer submitting a policy proposal to cosponsoring one.

Given these considerations, we adopt two strategies in identifying policy coalitions in the NPC and CPPCC. For the NPC, our interviews reveal an unspoken norm in NPC bill sponsorship: cross-province cosponsorship of a bill is discouraged and sometimes even prohibited. Hence, a typical NPC proposal is cosponsored by delegates from one province. To circumvent this unspoken norm during policy coalition building, some NPC delegates from different provinces submit similar bills, and sometimes these bills even share the same title. Informed by this insight, we adopt two measures of policy coalition in the NPC. The first measure of policy coalition includes only the bills that share exactly the same title in a given year. The second measure includes bills that have very similar titles but differ by a few words, which could be less precise but includes more bills.

Panel A in Figure 2 illustrates the pattern of policy coalition in the NPC by using the first measure. If a NPC bill is not part of the policy coalition, we should expect the following pattern: the average number of provinces behind a bill should be around one (i.e., the bill is sponsored by delegates from only one province), and the frequency of provinces joining the policy coalition should be zero (i.e., no additional province for a bill). Indeed, we observe many NPC bills follow this pattern in some NPC sessions. Nonetheless, we observe two cycles of policy coalitions, one in the 1980s and the other in the 2000s, and policy coalition building was significantly higher in the 2000s, peaking in 2005. In this graph, we also marked the promulgation of key laws concerning compulsory education between 1983 and 2007. Notably, these cycles coincide with promulgation of two laws in 1986 and 2006.¹⁸ Hence, these patterns suggest significant efforts in policy coalition building prior to the promulgation of these two laws. Panel B in Figure 2 adopts the broad measure of policy coalition, and the pattern remains consistent with Panel A. In addition, we identify two small policy coalitions during the 1990s by using this broader measure, which also coincides with the promulgation of two laws concerning compulsory education in 1993 and 1995.¹⁹ We repeat the same exercise using the total number of delegates participated in the coalitions, and Panels C and D in Figure 2 largely corroborate the patterns that we observed in the analysis of province participation.

Figure 2.

Policy coalition in the NPC (1983-2007).

We then turn to the analysis of the CPPCC. Note that the CPPCC does not have the same unspoken norm that discourages delegates from cosponsoring a policy proposal across provinces; hence, we must adopt a different measure of policy coalition in the CPPCC. Again, we aim to identify coordinated efforts in proposal submission on a particular policy position. Although CPPCC policy proposals are more heterogeneous and rarely share the same title, our interviews reveal that a policy coalition can be observed in two policy issues. First, if a policy coalition is organized by the MOE, it tends to focus on national issues instead of regional issues, allowing the MOE to pressure the central government, especially the Ministry of Finance (MOF), for a broader financial support. Thus, our first measure of a CPPCC policy coalition (National Policy) is whether some policy proposals address a national issue instead of a local issue. Second, the MOE sometimes aims to pressure the central government openly for greater education spending by making explicit demands for financial resources. Our second measure of CPPCC policy coalition (Demand Central Government Financial Resource) includes proposals in which financial responsibility from the central government is openly demanded.

Figure 3 illustrates the total numbers of CPPCC proposals and cosponsors in these two policy coalitions based on these two measures as well as the years when four laws concerning compulsory education were promulgated, and three interesting patterns emerge. First, we observe similar patterns in the coalition of National Policy and the coalition of Demand Central Government Funding, and the first coalition contains more proposals than the second coalition in most CPPCC sessions. This is not particularly surprising largely because joining the coalition of National Policy is less risky than the coalition of Demand Central Government Financial Resource because the second coalition is a direct challenge to the MOF—a powerful agency in the central government. Second, the rise and fall of both coalitions mimic the patterns of coalition building that we observe in the NPC. In particular, a greater number of policy proposals have made financial demands upon the central government since 2000. Third, and most importantly, the peaks of both types of coalitions in the CPPCC largely coincide with the promulgation of laws dealing with compulsory education in 1986, 1993, 1995, and 2006.

Figure 3.

Policy coalition in the CPPCC (1983-2007).

Based on the patterns we observed in the NPC and CPPCC, we argue that the correlation between the promulgation of laws and the peaks of NPC and CPPCC coalitions are hardly accidents. As illustrated in these figures, we observe a rising number of related NPC bills and CPPCC proposals concerning a particular issue prior to the promulgation of these laws, and the coalition dissipates immediately after the promulgations. Hence, the rise and fall of bills and proposal submission indicates coordinated efforts to advocate certain policy agendas in China’s national assemblies.

Empirics 2: The Participation of Policy Coalitions

The patterns of bill and proposal submissions presented in the last section suggest only coordinated efforts among delegates during some legislative sessions. Another important aspect of our definition of policy coalition is the coalition participation could surpass the delegates with ties to a particular government agency. Our theoretical framework suggests that bureaucratic influence operates mainly through delegates with specific employment ties to the central government agencies. However, the intensity of coalition increases when delegates without employment ties to a particular government agency also participate in coalitions that advocate the interests of this government agency. To evaluate these claims, we analyzed the background information of the delegates who initiated the NPC bill and CPPCC proposal, and Table 1 below reports the summary statistics. Two important attributes of the delegates are worth noting. The first attribute, Central Work Unit, is coded 1 when the delegate’s current employment is under a central government agency instead of a local government agency. The second attribute, Education Sector Tie, is coded 1 when the delegate’s concurrent or previous employment is in the education sector, such as the education institutions or bureaucracies. If a delegate belongs to both groups, then the MOE should directly control the employment of this delegate; hence, this delegate should be more subject to its bureaucratic influence. For example, almost all the higher education institutions are directly controlled by the MOE, so most universities are considered central government work units. Thus, the MOE could choose to intervene in promotions or transfers of any employees in higher education institutions, especially administrators.

Table 1.

The Delegate Characteristics of NPC Bill and CPPCC Proposal on Compulsory Education (1983-2007).

	NPC Bill		CPPCC Proposal
	M	N	M	N
Central work unit	23.2%	194	28.0%	1,220
Education sector tie	60.5%	190	62.5%	1,215
Female	32.8%	201	23.1%	1,221
Years of experience in national assemblies	4.29	199	5.17	1,208
CCP party member	59.2%	169	22.7%	1,186
Non-CCP party member	36.7%	169	53.6%	1,186
No party	4.1%	169	23.7%	1,186

Authors’ database, which includes 210 NPC bills and 1,255 CPPCC policy proposals on compulsory education between 1983 and 2007. Due to missing data problem, we were not able to collect the delegate characteristics for a few NPC bills and CPPCC proposals. NPC = National People’s Congress; CPPCC = Chinese People’s Political Consultative Conference; CCP = Chinese Communist Party.

Table 1 shows that 23% of the NPC bills and 28% of the CPPCC proposals concerning compulsory education were submitted by delegates with central government employment ties during the period from 1983 to 2007. Furthermore, 61% of the NPC bills and 63% of the CPPCC proposals were submitted by delegates with education employment ties. Finally, 17% of the NPC bills and 20% of the CPPCC proposals were submitted by those who have both central government employment ties and education employment ties.²⁰

The overall descriptive statistics reported in Table 1 cannot reveal the dynamics of coalition building across years. Figure 4 reports the percentages of NPC bills and CPPCC proposals submitted by delegates with employment ties to the MOE in each year, and several patterns emerge.²¹ First, the percentage of delegates with ties to the MOE is higher in the NPC than in the CPPCC, reflecting the tighter control of legislative activities in the NPC. In particular, because the CPPCC is more subject to bureaucratic influence because of the delegate selection process described above, we observe a greater percentage of CPPCC proposals concerning compulsory education submitted by delegates without any education ties. Second, we observe the percentages of delegates with employment ties to the MOE is lower in years (e.g., 1980s and 2000s) when we observe more intensified policy coalitions in both the NPC and the CPPCC depicted in Figures 2 and 3. Hence, this pattern provides evidence that that policy coalition building was intensified through the participation of outsiders. Third, policy coalition building intensified more in the 2000s than previous years, suggesting a new trend in which policy coalition building in China’s national assemblies involves a greater number of agency outsiders.

Figure 4.

The education tie of delegates (1983-2007).

Policy Coalition Building Through Network Mediation and Brokerage

To this point, we have shown policy coalitions exist in China’s national assemblies and the participation in the policy coalition involves both delegates with and without employment ties to the MOE. However, these patterns could be driven by the bandwagon effect among like-minded delegates without genuine efforts in policy coalition building. In this section, we combine network analysis with our interviews to demonstrate the ways through which delegates build policy coalitions in the NPC and CPPCC as well as the consequences of these policy coalitions.

Network analysis has been widely used by scholars to understand the dynamics of legislative activities. Studies have shown that network formation is based on institutional connections (Koger, Masket, & Noel, 2009; W. K. T. Cho & Fowler, 2010) and informal connections (Masket, 2008; Montgomery & Nyhan, 2017). Our interviews provide an illuminating insight that these two types of connections facilitate delegate recruitment during the coalition building process.²² The institutional connection occurs through party affiliation, because our interviews reveal that party discipline and coordination help a delegate recruit other delegates within the same party. In particular for delegates from non-CCP parties, we found that they often organize meetings and coordinate their bills and policy proposals prior to the annual NPC and CPPCC meetings. The informal connection is through geographical location of the employment. We argue that the proximity in geographical locations allows some delegates to actively interact with each other and persuade them to submit bills and proposals.

To illustrate how party and geographical connections facilitate policy coalition building, we employed network analysis to investigate CPPCC policy proposals. Our network analysis primarily focuses on the CPPCC for three reasons. First, the CPPCC is subject to less control by the CCP than the NPC; thus, we observe stronger policy coalition building in the CPPCC. Second, CPPCC delegates are selected by central elites without competitive elections, so they are more susceptible to bureaucratic influence. Finally, our CPPCC data contain far greater proposals than NPC bills for our network analysis.

Based on the characteristics of the delegates who initiated the CPPCC proposal, we mapped their connections through party and location ties for every year between 1983 and 2007. We used the Fruchterman–Reingold layout algorithm to draw the undirected network such that the distance between delegates is proportional to the shortest path linking them. The figures for all 25 yearly CPPCC networks are reported in Online Appendix 3 (Figures 1-25). Each figure presents two networks for the policy coalitions of National Policy and Demanding Central Government Financial Resources, two policy coalitions defined in the last section. In each network, the nodes represent delegates who submitted a policy proposal for a coalition in a given year, and the color and shape of the nodes indicate the employment ties with central government agencies and the education sector. The linkages between delegates are based on both party affiliation and provincial location of their employment.

Because of space limitations, we focus our discussion on three critical periods of policy coalition building behind the promulgation of education laws in 1986, 1995, and 2006. We corroborate our findings with insights from our interviews.

Policy Coalitions Building Between 1984 and 1986

We start with 1984, which marked the onset of the policy coalition building in the 1980s (Figure 5: column 1). As shown, the coalition of National Policy was organized by delegates with employment ties to the MOE, who used their party connections (CCP) and location connections (Beijing) to build policy coalitions. The momentum continued in the coalition of National Policy and extended to the coalition of Demanding Central Governance Financial Resource in the following year as these types of delegates remain the key players in 1985. These network patterns are consistent with our interviews. One senior MOE official (Interview ID: BJ10302012) emphatically told us that the MOE always tries to persuade NPC and CPPCC delegates to submit bills and policy proposals on their behalf, especially those who have employment ties with the MOE. This revelation was echoed by another interviewee (Interview ID: BJ12202012), who served in various working groups that drafted education-related laws.

Figure 5.

CPPCC policy coalition network between 1984 and 1986.

In 1986, the policy coalition building intensified further than in the previous two years as shown Figure 5, column 3. We observe a large cluster of non-CCP delegates participating in the policy coalition of National Policy, and two delegates mobilized other delegates not only within their own party (Jiu San Society, or Jiusan Xueshe) but also at their own provincial location (Beijing). The Jiu San Society consists mostly of scientists and university professors who work in academic institutions. These two delegates were important brokers, who connected different subgroups of delegates both vertically (i.e., other local delegates) and horizontally (i.e., other central government delegates). Although all these delegates worked in the higher education sector, their proposals were concerned about compulsory education.

Policy coalition building from 1984 to 1986 contributed to the promulgation of the Compulsory Education Law of the People’s Republic of China on April 12, 1986, which was the first specific legislation concerning compulsory education in the People’s Republic of China. This law emphasized safeguarding education funding by the central and local governments and decentralized the management of compulsory education to local governments through the “organizing education by three-level, managing education by two-level” policy (sanji banxue, liangji guangli). Markedly, the coalition of Demanding Central Governance Financial Resource was not as successful as the coalition of National Policy; hence, the central government was not obligated to provide financial support under this new arrangement. Although the 1986 Law stipulated the growth of education spending should be higher than the growth of fiscal revenues, it opened the door for local governments to collect additional education-related fees to finance compulsory education.

Policy Coalitions Building Between 1992 and 1995

As discussed in our theoretical framework, one of the objectives of building policy coalitions in the national assemblies is to solve the credible commitment problem in authoritarian politics. Since the decentralization of financing compulsory education to local governments in 1986, teacher compensation has been a major issue in CPPCC proposals since the late 1980s. Although the central government has released a number of policy directives to enhance grassroots education personnel’s compensation,²³ shortfalls in local education finance remained widespread across China.

The MOE has sought to establish greater fiscal commitment from the central government to safeguard compulsory education finance, but it often meets with opposition from other ministries (Interview ID: BJ12202012). For example, the MOE had intended to tie the hands of the MOF by inserting a “4 percent” education spending target²⁴ into the Education Law in 1995; nonetheless, the “4 percent” target has met strong resistance from the MOF. Eventually, the “4 percent” target was watered down to the “Three-Growth Policy” (“Sange Zengzhang”)²⁵ in the revision of the 1995 Education Law. The “Three-Growth Policy” was an unfunded mandate that pressured provincial governments to invest more in education spending, but the MOE was unsuccessful in building a policy coalition in the NPC because few delegates were willing to apply additional pressure provincial governments.

The strong opposition from the MOF helps explain the pattern of coalition building in the CPPCC in mid-1990s, which was not as intensive as the policy coalitions in the 1980s and 2000s. As shown in Figure 6, the policy coalitions of Demanding Central Governance Financial Resource between 1993 and 1995 were not as dense as the network in the 1980s. More importantly, delegates with direct employment ties to the MOE did not play an important role in either participation of the coalitions or bridging different networks of delegates together.

Figure 6.

CPPCC policy coalition network between 1993 and 1995.

Policy Coalitions Building Between 2002 and 2006

Figures 2 to 4 have shown that the policy coalitions, greatly intensified since 2000, eventually led to the revision of the Compulsory Education Law that established formal intergovernmental transfers from the central government to local governments that help finance compulsory education. The success of gaining fiscal commitment from the central government was a consequence of multiyear efforts in policy coalition building. Our interviews provided many important insights consistent with the patterns exhibited in our network analysis in Figure 7.

Figure 7.

CPPCC policy coalition network between 2002 and 2006.

An important insight of policy coalition building was revealed by a NPC delegate (Interview ID: HB03112013), who detailed his involvement in the process. This NPC worked in a higher education institution and was interested in promoting free compulsory education. Prior to the official 2003 NPC meeting, this newly elected NPC delegate was approached by a MOE official, who recommended he submit NPC bills concerning the revision of the Compulsory Education Law. This MOE official told the delegate that he has also contacted another NPC delegate from another province to submit similar bills. One major agenda was that the MOE would like to pressure the central government to provide greater education funding, and free compulsory education could be a good reasoning for requesting greater funding. Given the efforts by these two delegates as well as efforts by other delegates that they mobilized, more than 400 NPC delegates from four provinces cosponsored submitted or cosponsored bills to request a revision of the Compulsory Education Law and an increase in compulsory education finance in 2003. This delegate also mobilized other key CPPCC delegates through his party network, eventually leading more than 50 CPPCC proposals that were cosponsored by around 250 CPPCC delegates who requested greater fiscal demand from the central government in the same year. This delegate even personally delivered his proposal in writing to Premier Wen Jiapao when he visited the NPC meeting of this delegate’s provincial delegation.

Although the central government did not act on these bills and proposals immediately, this delegate cooperated with China Education, the newspaper that serves as the mouthpiece of the MOE, and organized a meeting among NPC and CPPCC delegates to continue to advocate the issues of free compulsory education and greater fiscal commitment from the central government.²⁶

China Education later published the major contents of this meeting, thus significantly generating media exposure across China. Moreover, the China Central Television (CCTV) interviewed this delegate to discuss these bills and proposals to advocate free compulsory education.

As revealed in our interviews, a successful policy coalition requires the mobilization of key political actors to build the momentum, and the coalition building often involves multiyear efforts. As shown in Figure 7, not only did a greater number of delegates participate in submitting education-related proposals, but several subgroups of delegates, who have no employment ties with the education sector, were also mobilized by key brokers through the party as well as the location connections. In 2003, for example, the leading non-CCP parties who mobilized their members are the Jiu San Society (Jiusan Xueshe), the China Association for Promoting Democracy (Minjin Party), the China Democratic League (Minmeng Party), and the China Democratic National Construction Association (Minjian Party). Note that the China Democratic League consists primarily of members working in education and culture in China, but the China Democratic National Construction Association comprises business elites. Hence, the mobilization spread to other noneducation-related groups.

By and large, the pattern exhibited in the network analysis is consistent with our interview of this delegate, who used both party and location connections, important venues for coalition building in this case. For example, this delegate belongs to the China Association for Promoting Democracy (Minjin Party), which was one of the major parties participating in the policy coalition. The effectiveness of party network for coalition building occurs because coordination among party members prior to NPC and CPPCC meetings is highly organized within the party as revealed by a senior official in one of the democratic parties in China (Interview ID: BJ02132015). The party always reviews the potential CPPCC proposals that delegates plan to submit. It also decides which issues to focus on during each CPPCC session. For those delegates who comply with the party, the party chairperson helps advance their careers by contacting high-level officials.

More importantly, we observe successful coalition building for both policy coalitions of National Policy and Demanding Central Government Financial Resource. By contrast, the coalitions of Demanding Central Government Financial Resources were not as successful in the 1980s and 1990s. The different outcomes of the Demanding Central Government Financial Resource coalition were in part driven by an important change in the Chinese fiscal system in 1999—the establishment of formal intergovernmental transfer channels from the central government to local government.²⁷ Prior to 1998, the central government made very few institutionalized intergovernmental fiscal transfers to local governments. Most intergovernmental transfers were ad hoc, targeting infrastructure spending. The establishment of institutional channels for intergovernmental transfers provides justification for the MOE and local governments to demand greater financial resources from the central government to finance local education.

Success in building the coalition of Demanding Central Government Financial Resource continued after 2003. Not only did we observe more intensity in NPC coalitions since 2003, but we also observed more intensified participation by CPPCC delegates. The policy coalition eventually paved the way for the revision of the Compulsory Education Law of the People’s Republic of China on June 29, 2006. The revised 2006 Law pinpoints the financial responsibilities of the central government for compulsory education, and mandates the expansion of fiscal transfers from central and provincial governments to local governments to finance compulsory education.²⁸ Indeed, the State Council established a set of policies in the name of the “New Mechanism” (xin jizhi), which significantly increased intergovernmental transfers to local governments.²⁹ As a result of this reform, education spending as a percentage of GDP steadily increased from 2005, and finally reached 4 percent of the GDP in 2012.

Network Centrality Analysis

This network analysis offers important insights about the key role played by the MOE-controlled delegates in coalition building in the CPPCC. One may argue that the illustrations of networks are only descriptive and subject to different interpretations. To formally analyze these networks, we calculated centrality measures to identify the key delegates in each network. The first network centrality indicator is betweenness, which identifies a node’s position within a network in terms of its ability to make connections to other pairs in a network. This indicator allows us to identify the delegate who has the most influence in connecting other delegates who otherwise would not have been connected. A second network centrality measure is eigenvalue, an indicator measuring the degree to which a node is connected to another highly connected node. This indicator allows us to capture the ability of a delegate to connect with other influential delegates who also have high mobilization capacity with many connections. Finally, we follow Gould and Fernandez (1989) by calculating the brokerage measure, which allows us to identify the key delegates who connect various subgroups of delegates instead of individual delegates.

Table 2 reports the summary statistics of these centrality measures in CPPCC policy coalitions. We find strong evidence that delegates with employment ties to the MOE played a critical role in bridging delegates in policy coalitions. Across all network centrality measures, these delegates have higher scores than other delegates, and the differences are statistically significant. Taken together, our results suggest that employment ties to the MOE are critical in creating bridges between delegates as well as connecting influential delegates who have greater mobilization capacity themselves. Substantively, these results corroborate our argument that delegates must signal their loyalty to central government agencies; thus, they are more active not only in bill initiation but also in building policy coalitions in the national assemblies.

Table 2.

Network Analysis of Policy Coalition.

	Coalition of national policy			Coalition of demanding central government financial resources
	Betweenness	Eigenvalue	Brokerage	Betweenness	Eigenvalue	Brokerage
	(1)	(2)	(3)	(4)	(5)	(6)
Direct MOE Employment Tie = 1
M	0.035	0.566	−0.592	0.034	0.532	−0.339
SE	(0.004)	(0.027)	(0.124)	(0.007)	(0.041)	(0.161)
n	207	207	207	91	91	91
Direct MOE Employment Tie = 0
M	0.023	0.306	−1.272	0.024	0.301	−0.997
SE	(0.001)	(0.013)	(0.046)	(0.002)	(0.019)	(0.062)
n	685	685	685	347	349	345
Significant difference level	0.003***	0.000***	0.000***	0.138	0.000***	0.000***

We report the means of network centrality measures between the group of delegates with direct employment ties to the MOE and otherwise. We coded direct MOE employment tie to 1 when the delegate’s employment is linked to both the central government agency and education sector. All the network centrality measures are standardized. T test results are based on the assumption of unequal variances. MOE = Ministry of Education.

p < .1. **p < .05. ***p < .01.

Conclusion

Studying authoritarian institutions has renewed interest among scholars in recent years; yet, most studies focus on their functions to maintain regime survival. As demonstrated in our research, studying legislative activities of authoritarian national assembly could offer new insights into authoritarian politics and policy making. Employing a new analytical lens, we argue that authoritarian legislatures could serve as an important venue among bureaucracies to advance their agenda in policy making. Based on an original dataset of education-related bills and proposals in China’s national assemblies between 1983 and 2007, we show that bureaucratic influence leads to policy coalition building in the NPC and CPPCC despite the control of these institutions by the CCP. The success of policy coalition building resulted in changes in laws and, more importantly, increases in financial resources for compulsory education.

Our article contributes to the scholarship that emphasizes the importance of bargaining among government bureaucracies in policy making in China (Lieberthal & Lampton, 1992; Lieberthal & Oksenberg, 1988; Mertha, 2009). We show that bureaucratic bargaining migrates from private domain (i.e., closed-door meetings) to public domain (i.e., the NPC and CPPCC). As a result, legislative activism is not necessarily confined to the local congresses in China. One important new insight from our research is that legislative activism exists even in the national assemblies and that is through building policy coalitions via bills and proposals. Our finding is particularly relevant to a class of authoritarian regimes whose political systems produce persistent structural conflicts among different functioning groups. In this institutional setting, legislative behavior reflects strategic interaction and coalition building among different groups of elites.

This article offers three important implications for understanding the politics of authoritarian legislatures. First, these kinds of so-called democratic institutions are still driven primarily by elite politics, especially when the electoral connection to the masses is weak or absent. Nonetheless, the proliferation elite bargaining from the closed-door meeting to the more visible national assembly suggest an important development in authoritarian politics. Understanding the condition under which this political development occurs provides fruitful insights into authoritarian politics. Second, the involvement of delegates without employment ties in a policy coalition indicates a potential pattern of logrolling behaviors among delegates in Chinese national assemblies. Investigating coalition building beyond the education sector could point to a fruitful future research direction to understand the political dynamics of authoritarian legislature. Our observation of legislative activities in other sectors suggests that coalition building by central government agencies is not unique in the education sector. For example, the recent development of legislation that supports civil lawsuits against the government was pushed mainly by the judicial system at the central government, which allows them to constrain local government behaviors. Another illuminating example is the recent political activism by the military system that has lobbied for greater compensation for discharged and retired military personnel, who obtain their benefits mainly from local governments. We observe that a greater number of NPC delegates without ties to the military delegation have been mobilized to participate in these coalitions since 2013. If logrolling behavior becomes a common practice in authoritarian legislatures, it could significantly shape the policy making process dominated by central elites.

Finally, one important scope condition of our argument is that it is unlikely to observe policy coalitions in sensitive policy areas. Studying the NPC in China, Truex (2016) coined the term “representation within bound,” as he suggests that NPC delegates restrain themselves from pressing the government in more sensitive policy areas such as political reforms and human rights issues. Notably shown in our data, the intensity of policy coalition building varies across time and has been more prominent in recent years. In addition, the policy coalitions do not always lead to successful policy outcomes. We believe the cyclical patterns and the policy areas of coalition building in authoritarian legislatures are in part driven by the dynamics of the balance of power among ruling elites and inter-elite bargaining. This is an important area worth investigation in the future.

Supplemental Material

CPS797950_Online_appendices – Supplemental material for Policy Coalition Building in an Authoritarian Legislature: Evidence From China’s National Assemblies (1983-2007)

Supplemental material, CPS797950_Online_appendices for Policy Coalition Building in an Authoritarian Legislature: Evidence From China’s National Assemblies (1983-2007) by Xiaobo Lü, Mingxing Liu and Feiyue Li in Comparative Political Studies

Supplemental Material

CPS_Replication_Files – Supplemental material for Policy Coalition Building in an Authoritarian Legislature: Evidence From China’s National Assemblies (1983-2007)

Supplemental material, CPS_Replication_Files for Policy Coalition Building in an Authoritarian Legislature: Evidence From China’s National Assemblies (1983-2007) by Xiaobo Lü, Mingxing Liu and Feiyue Li in Comparative Political Studies

Footnotes

Acknowledgements

We received helpful comments from Ben Ansell, Pablo Beramendi, Jason Brownlee, John Bullock, Martin Dimitrov, Anna Grzymala-Busse, Yue Hou, Wendy Hunter, Franziska Keller, Jennifer Gandhi, Kenneth Greene, Eddy Malesky, Melanie Manion, Ciqi Mei, Ben Noble, Thomas Remington, Paul Schuler, Milan Svolik, Rory Truex, Yuhua Wang, Rachel Wellhausen, Matthew Wilson, and Xufeng Zhu as well as panel and seminar participants at Columbia University, Duke University, Fudan University, Hoover Institution, Tsinghua University, University of California at Berkeley, University of California at San Diego, University of Michigan, University of Oxford, University of Texas at Austin, the 2014 and 2015 American Political Science Association annual conference, and the 2015 Midwest Political Science Association annual conference for various versions of this article. Mickey Li has provided valuable research assistance. All errors remain our own.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Supplemental Material

Supplemental material for this article is available online at the CPS website

Notes

Author Biographies

Xiaobo Lü is an associate professor in the Department of Government, the University of Texas at Austin. His research focuses on comparative political economy and Chinese politics, and is published in the American Political Science Review, American Journal of Political Science, Comparative Political Studies, and other outlets.

Mingxing Liu is a professor of economics at the China Institute of Educational Finance Research at Peking University in China. His research interests include political economy of development, fiscal policies, and local governance. He has published articles in journals such as American Political Science Review, Comparative Politics and China Quarterly.

Feiyue Li received his doctorate degree in development economics from Peking University in June 2009. He is a native of China and holds a bachelor’s degree in law from Zhejiang University, and a master’s degree in political economy from Nankai University. His research interests include the political economy of development and economic growth.

References

Adams

D’Onofrio

Sokoloff

(1981). The iron triangle: The politics of defense contracting. New York, NY: Council on Economic Priorities.

Blaydes

(2010). Elections and distributive politics in Mubarak’s Egypt. New York, NY: Cambridge University Press.

Boix

Svolik

M. W.

(2013). The foundations of limited authoritarian government: Institutions, commitment, and power-sharing in dictatorships. The Journal of Politics, 75, 300-316.

Brownlee

(2007). Authoritarianism in an age of democratization. Cambridge, UK: Cambridge University Press.

Cho

W. K. T.

Fowler

J. H.

(2010). Legislative success in a small world: Social network analysis and the dynamics of congressional legislation. The Journal of Politics, 72, 124-135.

Cho

Y. N.

(2009). Local people’s congresses in China: Development and transition. Cambridge, UK: Cambridge University Press.

Chu

Y. H.

Lin

J. W.

(2001). Political development in 20th-century Taiwan: State-building, regime transformation and the construction of national identity. The China Quarterly, 165, 102-129.

Desposato

S. W.

(2001). Legislative politics in authoritarian Brazil. Legislative Studies Quarterly, 26, 287-317.

Dimitrov

M. K.

(2014). Internal government assessments of the quality of governance in China. Studies in Comparative International Development, 50, 50-72.

10.

Eggers

A. C.

Hainmueller

(2009). MPs for sale? Returns to office in postwar British politics. The American Political Science Review, 103, 513-533.

11.

Fenno

R. F.

(1978). Home style: House members in their districts. New York, NY: HarperCollins.

12.

Gallagher

M. E.

Hanson

J. K.

(2015). Power tool or dull blade? Selectorate theory for autocracies. Annual Review of Political Science, 18, 367-385.

13.

Gandhi

(2008). Political institutions under dictatorship. New York, NY: Cambridge University Press.

14.

Gandhi

Lust-Okar

(2009). Elections under authoritarianism. Annual Review of Political Science, 12, 403-422.

15.

Gould

R. V.

Fernandez

R. M.

(1989). Structures of mediation: A formal approach to brokerage in transaction networks. Sociological Methodology, 19, 89-126.

16.

Hou

(2015, April). Participatory autocracy: Private entrepreneurs, legislatures, and property protection in China. 2015 Midwest Political Science Association Annual Conference, Chicago.

17.

Hough

J. F.

Fainsod

(1979). How the Soviet Union is governed. Cambridge, UK: Harvard University Press.

18.

Huang

Boranbay-Akan

Huang

(2019). Media, protest diffusion, and authoritarian resilience. Political Science Research and Methods, 7(1), 23-42.

19.

Kelley

D. R.

(1972). Interest groups in the USSR: The impact of political sensitivity on group influence. The Journal of Politics, 34, 860-888.

20.

Kessler

Krehbiel

(1996). Dynamics of cosponsorship. The American Political Science Review, 90, 555-566.

21.

King

Pan

Roberts

M. E.

(2014). Reverse-engineering censorship in China: Randomized experimentation and participant observation. Science, 345(6199), 1-10.

22.

Koger

Masket

Noel

(2009). Partisan webs: Information exchange and party networks. British Journal of Political Science, 39, 633-653.

23.

Lieberthal

Lampton

D. M.

(1992). Bureaucracy, politics, and decision making in post-Mao China. Berkeley: University of California Press.

24.

Lieberthal

Oksenberg

(1988). Policy making in China: Leaders, structures, and processes. Princeton, NJ: Princeton University Press.

25.

Lorentzen

(2013). Regularizing rioting: Permitting public protest in an authoritarian regime. Quarterly Journal of Political Science, 8, 127-158.

26.

Lorentzen

(2014). China’s strategic censorship. American Journal of Political Science, 58, 402-414.

27.

MacFarquhar

(1998). Provincial people’s congresses. The China Quarterly, 155, 656-667.

28.

Magaloni

(2006). Voting for autocracy: Hegemonic party survival and its demise in Mexico. New York, NY: Cambridge University Press.

29.

Mahdavi

(2015). Explaining the oil advantage: Effects of natural resource wealth on incumbent reelection in Iran. World Politics, 67, 226-267.

30.

Malesky

Schuler

(2010). Nodding or needling: Analyzing delegate responsiveness in an authoritarian parliament. American Political Science Review, 104, 482-502.

31.

Malesky

Schuler

(2011). The single-party dictator’s Dilemma: Information in elections without opposition. Legislative Studies Quarterly, 36, 491-530.

32.

Manion

(2008). When communist party candidates can lose, who wins? Assessing the role of local people’s congresses in the selection of leaders in China. The China Quarterly, 195, 607-630.

33.

Manion

(2014). Authoritarian parochialism: Local congressional representation in China. The China Quarterly, 218, 311-338.

34.

Manion

(2015). Information for autocrats: Representation in Chinese local congresses. Cambridge: Cambridge University Press.

35.

Masket

S. E.

(2008). Where you sit is where you stand: The impact of seating proximity on legislative cue-taking. Quarterly Journal of Political Science, 3, 301-311.

36.

Mayhew

D. R.

(1974). Congress: The electoral connection. New Haven, CT: Yale University Press.

37.

Mertha

(2009). “Fragmented authoritarianism 2.0”: Political pluralization in the Chinese policy process. The China Quarterly, 200, 995-1012.

38.

Miller

M. K.

(2015). Elections, information, and policy responsiveness in autocratic regimes. Comparative Political Studies, 48, 691-727.

39.

Montgomery

J. M.

Nyhan

(2017). The effects of congressional staff networks in the US House of Representatives. The Journal of Politics, 79, 745-761.

40.

Myerson

R. B.

(2008). The autocrat’s credibility problem and foundations of the constitutional state. American Political Science Review, 102, 125-139.

41.

Noble

(2017). Amending budget bills in the Russian State Duma. Post-Communist Economies, 29, 505-522.

42.

O’Brien

K. J.

(1990). Reform without liberalization: China’s National People’s Congress and the politics of institutional change. New York, NY: Cambridge University Press.

43.

O’Brien

K. J.

(1994). Chinese People’s Congresses and legislative embeddedness: understanding early organizational development. Comparative Political Studies, 27, 80-107.

44.

Remington

T. F.

(2001). The Russian parliament: Institutional evolution in a transitional regime, 1989-1999. New Haven, CT: Yale University Press.

45.

Reuter

O. J.

Robertson

G. B.

(2015). Legislatures, cooptation, and social protest in contemporary authoritarian regimes. The Journal of Politics, 77, 235-248.

46.

Shih

V. C.

(2008). Factions and finance in China: Elite conflict and inflation. New York, NY: Cambridge University Press.

47.

Shirk

S. L.

(1993). The political logic of economic reform in China. Berkeley: University of California Press.

48.

Skilling

H. G.

(1966). Interest groups and communist politics. World Politics, 18, 435-451.

49.

Skilling

H. G.

Griffiths

(1971). Interest groups in Soviet politics. Princeton, NJ: Princeton University Press.

50.

Skilling

H.G.

(1983). Interest groups and communist politics revisited. World Politics, 36, 1-27.

51.

Svolik

M. W.

(2012). The politics of authoritarian rule. New York, NY: Cambridge University Press.

52.

Tanner

M. S.

(1995). How a bill becomes a law in China: Stages and processes in lawmaking. The China Quarterly, 141, 39-64.

53.

Tanner

M. S.

(1999). The politics of lawmaking in post-Mao China: Institutions, processes, and democratic prospects. Oxford, UK: Oxford University Press.

54.

Truex

(2014). The returns to office in a “rubber stamp” parliament. American Political Science Review, 108, 235-251.

55.

Truex

(2016). Making autocracy work: Representation and responsiveness in modern China. New York, NY: Cambridge University Press.

56.

Wang

(2016). Beyond local protectionism: China’s state-business relations in the last two decades. The China Quarterly, 226, 319-341.

57.

Xia

(2008). The people’s congresses and governance in China: Toward a network mode of governance. London, England: Routledge.

58.

Yang

(2004). Remaking the Chinese Leviathan: Market transition and the politics of governance in China. Stanford, CA: Stanford University Press.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.07 MB

0.00 MB

0.59 MB