Abstract
Recent studies have begun to pay increasing attention to congressional representation in China. Based on selected cases or surveys of one province or another, such studies seek to demonstrate that the deputies of local people’s congresses (LPCs) increasingly identify themselves as representatives of citizens rather than as state agents. This article, using data from a national survey conducted in 2014, explores how deputies at the county level perceive their role and what accounts for their different role perceptions. It argues that LPC deputies as a whole perceive that they have overlapping roles that could be defined as neither state agents nor citizen representatives. Rather, they try to strike a balance between seemingly contradictory roles. Deputies’ social background and political attitudes have a significant effect on their role perceptions while electoral incentives make little difference. A sense of congressional representation develops when deputies gradually come to grips with the tension between the different roles and choose to give up roles other than citizen representatives.
Since China does not have competitive elections, researchers have long considered the deputies of China’s people’s congresses as agents of the state instead of citizen representatives. Scholars have argued that strengthening ties with constituents is less important for deputies than strengthening ties with superiors; their mandate is from above instead of from below (O’Brien and Luehrmann, 1998). Hence, most studies have focused on the agency of the people’s congresses rather than the deputies. They contend that China’s local people’s congresses (LPC) are embedded in the party-state and adopt subtle tactics that allow them to supervise the government and the courts, even though they enjoy limited autonomy and their deputies lack incentives to speak for their constituencies (O’Brien, 1994b; Xia, 1997, 2000, 2007; Cho, 2002, 2003a, 2003b). However, more recently, some studies have begun to look more closely at the issue of representation in China (see, e.g., Manion, 2008). They seek to demonstrate that LPC deputies identify themselves increasingly as representatives of ordinary citizens and are more likely to advocate on their behalf (Manion, 2014). These studies, valuable though they be, disagree on two critical questions. First, whom do deputies think they represent? And, second, what explains the differences between deputies’ sense of representation?
Given that recent studies are largely based on either selected cases or survey data from a single province, the disagreement is partly rooted in selection bias. Melanie Manion’s latest study partly relies on data from original probability sample surveys of 5,130 local congress deputies across three provinces: Anhui, Hunan, and Zhejiang. Her study is very inspiring and insightful, but her questionnaire does not use questions about political attitudes and electoral incentives as independent variables. This article uses the latest national sampling survey data and seeks to answer the two questions mentioned above. The survey was carried out from September 2013 to January 2014. The section below under the heading “Sampling Scheme” describes the survey. As the sense of representation varies according to the level of the congress involved and since deputies in lower-level congresses have been considered to be more responsive to constituents’ needs and complaints than those in higher-level congresses (Manion, 2009), our survey focuses on county-level deputies who are directly chosen by the electorate.
We look closely at deputies’ perception of their role, which should not be confused with the actual fulfillment of their role. Based on the latest national survey, we argue that LPC deputies as a whole have a distinctive sense of representation, in the sense that they try to strike a balance between seemingly contradictory roles. First, as a whole, deputies have an overlapping role perception, which could be characterized as being neither a state agent nor a citizens’ representative. Most deputies at the county level do not want to challenge the government, but choose to stand with ordinary people when government policy clashes with the public interest. They tend to consider themselves as acting on behalf of ordinary citizens, but some of them also expect to derive personal honor and benefits from their job. Moreover, they are likely to agree with either the mandate or the trustee view of representation (discussed below). Second, deputies’ social background and political attitudes have a significant effect on their role perceptions, while electoral incentives—that is, the desire to be reelected—make little difference. Deputies who strongly support the regime and who have authoritarian values are more likely to play overlapping roles, while deputies with a particular employment background, such as those who have worked in nonpublic-owned enterprises, tend to be more sensitive to the contradictions in their role. Third, we argue that congressional representation develops when deputies gradually recognize the tension between different roles and increasingly choose to side with their constituency.
Nascent Congressional Representation in China
In a competitive election, the electoral connection is usually deemed the most important factor constraining legislators in representing their constituency. Congressmen are assumed to be single-minded seekers of reelection. Thus, the incentive of reelection decides the behavior of legislators and drives them to engage in activities to benefit their constituency (Mayhew, 1974; Bianco, Spence, and Wilkerson, 1996; Katz and Sala, 1996). Conversely, legislators are more likely to shirk their responsibilities if they plan to retire and want to free themselves of electoral constraints or if they pursue higher office and face a different constituency (Rothenberg and Sanders, 2000). Even if such legislators remain committed, they will pay less attention to their constituency, attend fewer roll-call votes, and instead focus more on legislative issues (Herrick, Moore, and Hibbing, 1994). Direct and narrow-margin elections also contribute to the electoral connection. Direct elections can intensify individual electoral incentives to respond to the appeals of one’s constituency (Meinke, 2008). In addition, according to the so-called marginality hypothesis, representatives who come from highly competitive districts and win elections by narrow margins are more sensitive to the electorate than legislators who hold safe seats (MacRae, 1952). It is not just the size of the margin in the preceding election that matters but also the subjective judgments of the possible size of the margin that affect the electoral connection (Fiorina, 1973; MacRae, 1952). Although some studies contend that the relationship between electoral competition and responsiveness is complex and nonlinear, they do not deny that competition encourages responsiveness (Campbell, 1981).
By contrast, scholars have viewed people’s congress deputies as regime insiders who pose little challenge to the regime even though they reflect public opinion and offer proposals to the government. Kevin O’Brien identifies three sub-roles deputies play: inactive, agent, and remonstrator. For O’Brien, remonstrators continue to fulfill their responsibilities as regime agents even though they pay more attention to the interests of their geographic constituency and different social groups. Confrontation between the government and remonstrators is by no means frequent and the greatest potential for deputies to exercise influence arises when sub-roles meet (O’Brien, 1994a). Roderick MacFarquhar also finds that deputies can play a role even though they are not elected by their constituents; in this case, the role of the deputy is similar to that of an ombudsman who receives complaints from the broad public and makes different but equally loyal suggestions to the government (MacFarquhar, 1998).
Recent studies seek to demonstrate that deputies nowadays are more active and have gradually developed a sense of representation that is reflected in both their attitudes and actions (Manion, 2009). These studies consider deputies as no longer primarily engaged with the responsibilities that come with being state agents. Instead, most deputies consider themselves representatives of ordinary people (Manion, 2014). Some other studies still hold different views on whom LPC deputies represent and what accounts for the different representations among deputies.
First, some scholars have argued that LPC deputies tend to represent a particular social class and region. Young Nam Cho contends that deputies nowadays act mainly as public supervisors, reflectors, and policy providers rather than “regime agents.” Since deputies are involved in a part-time representative system, some of them tend to care most about issues related to their daily work and social background rather than the interests of the electorate. They are inclined to view themselves as “a deputy for farmers,” “a deputy for private businessmen,” or “a deputy for women.” Even deputies who are officials consider themselves representatives of a particular area, such as public security, traffic, or the courts (Cho, 2003b). Second, other scholars have contended that LPC deputies have developed an increased sense of constituency and have become more likely to represent their geographic constituency. Manion uses a survey of LPC deputies in three provinces to show that only a small minority agrees that they should obey the decisions of the Chinese Communist Party (CCP) and that the CCP represents the interests of their constituency. Furthermore, LPC deputies do not just stand in for the public or a particular social class, rather, they have a sense of constituency, which is reflected in the following statement: “Congress delegates should side with (baochi yi zhi 保持一致) the majority of their constituents, because constituents best understand their own interests” (Manion, 2014: 321). Ninety-two percent of Manion’s respondents agreed or basically agreed with this mandate view of representation (Manion, 2015: 87, table 3.1). Manion’s data show that the responsibility to transmit downward resonates little among the delegates surveyed: fewer than 2% rank it as most important while more than 50% consider as their most important responsibility doing practical things for the public and solving concrete problems. As for activities, constituent contacting is more common than before and deputies tend to speak more for their geographic constituencies (Manion, 2015: 327). Therefore, as far as the LPC is concerned, the biggest component of representation is what Manion calls geographically parochial responsiveness (Manion, 2009, 2014). Third, deputies also tend to voice their own opinions in the name of ordinary citizens and even vie for a portion of the spoils for themselves. Although they have limited access to state patronage, deputies have become more aggressive than ever about articulating their personal interests. Studies of private entrepreneurs reveal that they sometimes make full use of their identity as a deputy to promote their personal interests (Gandhi and Lust-Okar, 2009; He, Tian, and Chen, 2007; Truex, 2014; Hou, 2015). Sometimes deputies are also inclined to pursue the interests of their friends, family, and acquaintances (Gandhi and Lust-Okar, 2009).
What, then, accounts for the nascent congressional representation in the LPC? Electoral competition has been strictly constrained in China. Elections for the people’s congresses are often considered a process of co-option to secure social compliance (Sun, 2014; Dickson, 2000, 2003, 2007). O’Brien believes that deputies may speak for the people, but not because the people control their fate (O’Brien, 1988). What might drive deputies to act on behalf of their constituencies? First, some studies argue that the limited elections in China do contribute to electoral connections because China’s electoral system is designed to align preferences of “selectorates” and electorates, and candidates have to win the support of the electorate (Manion, 2000). In village elections, leaders and ordinary villagers have similar viewpoints, this because of the electoral connection rather than the shared local environment, informal influence, or socialization (Manion, 1996). As for people’s congress elections, Manion argues that those elections also produce electoral incentives to bolster congressional representation because competition in an election is by no means absent and the party-state cannot guarantee an electoral victory, even though it seeks to control the election process. It is elections that contribute to LPC deputies’ nascent sense of representation, especially in township and county-level direct elections (Manion, 2014). According to this argument, voter nominees and party nominees are significantly different types. Voter nominees may share their constituents’ views about local problems; they are at least spatially implicated in them. They may also be more susceptible to informal community influences. Party nominees, on the other hand, tend to reflect officially valued competence and are presumed to be loyal to the regime. Another viewpoint also supports the electoral connection argument, but it places more emphasis on community pressure, not reelection, as promoting the electoral connection. In this view, as long as deputies are truly nominated and elected by their constituents through non-official means, they are more likely to be active participants in crafting legislation. However, reelection pressure is an unlikely explanation for the electoral connection because being a deputy is not a professional full-time political job. It is community (constituent) pressure from frequent interaction and close familiarity that presses deputies to speak for their constituency (Guo, 2007).
Second, deputies’ social background plays an important role in representation. Party affiliation has long been considered closely linked to the electoral connection. In Western democracies, representatives are likely to take a partisan stance rather than speak for their local constituency when they do not intend to seek reelection (Swift, 1987; Studlar and McAllister, 1994). Given that most of China’s LPC representatives are still selected by the party-state, party affiliation and state employment background might play an even more important role. Studies have found that official-deputies are more concerned with the survival of the whole regime and the interests of the ruling elite than the people’s interests (Xia, 2000). By contrast, independent candidates, who are actually nominated by voters, are largely not members of the CCP (He, 2010). Furthermore, it is noteworthy that deputies’ occupational background also seems to have an impact on their representation. Entrepreneurs, lawyers, and journalists are among those who are likely to be active, independent candidates (He, 2010).
Third, political attitudes can also affect congressional representation. The theory of the electoral connection views congressmen as rational, political actors motivated by electoral incentives. However, when we discuss congressional representation in China, political attitudes may be particularly important. Those who identify more with the current regime and those with authoritarian values consider the party-state as the source of their mandate and thus are less likely to view themselves as representatives of ordinary citizens. By contrast, our study finds that civic-minded deputies are more inclined to vigorously pursue constituent or group concerns despite there being few rewards for doing so and few sanctions for ignoring their constituents (Xia, 2007).
Building on these insightful and illuminating studies of legislators’ perception of their role, this article looks more deeply into four crucial aspects of this subject. First, it distinguishes role perception from actual role fulfillment. Recent studies consider deputies’ role perceptions as congruous with their actual role fulfillment, and thus presume that what accounts for role perception also accounts for role fulfillment. However, as Gabriel Almond and Sidney Verba pointed out, political behavior is substantially influenced by institutional settings and thus political attitudes cannot be inferred from political behavior (Almond and Verba, 1989). Nowadays deputies may act more in response to the concerns of their constituents, but this does not necessarily mean they are more likely to have a sense of representation or be more defiant than in past years. It is entirely possible that deputies with a strong sense of representation can be relatively inactive. This is because role fulfillment is relatively susceptible to electoral and institutional incentives, while role perceptions can be more complicated and influenced by other factors (see Huang and Chen, 2015). Thus, this article explores only the role perception of deputies. Second, we asked deputies seemingly contradictory questions about their role perceptions. Given that deputies would like to respond positively to statements that seem to them politically correct, we tried to include in our questionnaire all possible roles mentioned in the recent literature. Thus we asked deputies to indicate whether they agree with contradictory role perceptions. Third, the types of nominees in our questionnaire were defined according to the subjective perceptions of our respondents. Manion used national election data in which the deputies were categorized into party nominees and voter nominees. The latter refers to those who were nominated by a district constituency, including passive voter nominees, independent candidates, and pseudo-voter nominees (see Manion, 2015: 56-58). Her data analysis finds that elections contribute to representation in that party nominees and voter nominees are significantly different. However, since no more independent variables were controlled, it is hard to distinguish pseudo-voter nominees from other types of voter nominees. Nonetheless, it is still worth exploring whether an electoral connection does exist. Fourth, our analysis includes a greater number of important independent variables. Our regression model includes, among others, political attitudes, state employment background, and reelection incentives. In sum, based on our national survey data, our study makes a distinction between role perception and actual role fulfillment and explores how deputies at the county level perceive their role and what accounts for their different role perceptions.
Sampling Scheme
Our national survey of people’s congress deputies at the county level was carried out by a research group at Sun Yat-sen University from September 2013 to January 2014. Our research partners at five universities in China—Zhejiang University, Wuhan University, Jilin University, Guizhou University, and Guangxi University of Nationalities—were commissioned to implement the survey. Trained surveyors from the five universities handed out the questionnaires to county-level LPC deputies, with the permission and help of the LPCs. The supervisors from Sun Yat-sen University were in charge of the quality of the survey. The questionnaires were sent to deputies when the LPC was in session, when they were gathered for group activities, or when they were organized by the township-level people’s congress and the street people’s congress working committee. To prevent possible self-censoring, the questionnaires included statements to the effect that Sun Yat-sen University was conducting the survey for academic use and that no one else would have access to the questionnaires.
The data were collected by stratified three-stage PPS (probability proportionate to size) sampling. First, given that the number of LPC deputies in each county and province has not been officially disclosed, we estimated the number of county-level LPC deputies according to the deputy quota allocation scheme provided by the Organic Law of the National People’s Congress of the PRC. That is to say, the minimum number of deputies is 120 and the maximum 450, with one deputy per 5,000 people in each county. The population of each county is based on the 2010 national census. We added up the number of county-level LPC deputies in each county to get an estimated total of the deputies in each province.
Second, the provinces (and autonomous regions and municipalities) were divided into three geographic areas according to official standards: East China, Central China, and West China. In each area, stratified three-stage PPS sampling was carried out. In the first stage, with the provinces in each area as the first sampling unit (PSU), we sorted the PSUs based on provincial GDP (2011 data) in descending order. We drew two provinces from each area by systematic interval sampling based on the cumulative number of the county-level LPC deputies in each province. As a result, Zhejiang and Guangxi in East China, Hubei and Jilin in Central China, and Inner Mongolia and Guizhou in West China were selected. In the second stage, we sorted the counties (districts) in each selected province based on the proportion of the nonagricultural population in descending order and drew three counties/districts from each selected province with the same method of internal sampling based on the number of county-level LPC deputies in each county. In the third stage, we drew 70 deputies from each selected county and district using random sampling. It is noteworthy that the survey of the Inner Mongolia Autonomous Region could not be conducted because it was not possible for us to get approval from the LPC.
Third, we carried out interlayer and intralayer weighting. In the process of implementing the survey, we selected more than 70 samples in each county or district in case of refusal. In addition, in some counties or districts, the number of deputies is relatively small and is nearly equal to the sampling number, so we had to give questionnaires to all the deputies. Therefore, the actual sample included 1,187 cases. We needed to weight the cases in each county to ensure that the probability for each case to be selected was equal. The weighting value added to each case was calculated according to the formula: iw = 70 / ni, where ni refers to the actual number of samples collected in each county. Furthermore, as mentioned, we also conducted intralayer weighting. Since the total number of county-level people’s congress deputies in each area was different, and since the survey of the Inner Mongolia Autonomous Region failed for technical reasons, as we have explained, we also weighted the cases in each area to ensure that the probability of a deputy being selected in each area would be nearly equal. The weighting value is calculated based on the formula hw = qcg / npg, where qcg refers to the quota allocated to each area and npg refers to the actual number of samples collected in each area. In the end, according to the above analysis and calculation, the comprehensive weighted value was calculated based on the formula: tw = iw × hw (see Table 1).
Intralayer Weighting Value.
Role Perceptions of County-Level LPC Deputies
In our national sampling survey, we asked three sets of questions to uncover county-level LPC deputies’ role perceptions. As we can see from the results, the deputies do not consider seemingly contradictory role perceptions to be in conflict with each other. Rather, their role perceptions overlap and they try to strike a balance between the different roles.
Personal Honor and Benefit
The first set of questions is about whether being a people’s congress deputy can be a means to acquire personal honor and benefits. It includes five questions, as shown in Table 2. The respondents were asked to choose their answers based on a 5-point ordinal scale: 1 and 2 stand for strongly disagree and disagree, respectively; 3 stands for neither agree nor disagree; and 4 and 5 stand for agree and strongly agree, respectively. To ensure internal consistency, we tested reliability in the present sample for the scale. The Cronbach α coefficient is >0.844. The internal consistency is rather good. Next, these five items were combined to form an additive index. Table 2 shows the results. There is a mean score of 14.73 for the entire index, a little below the mid-point (15) of the index scale. Among the five items, the mean of the second statement (3.12), that it is helpful to make friends in order to be selected as a people’s congress deputy, ranks highest. By contrast, the mean of the fifth statement, that it is helpful to establish connections with the government officials, ranks lowest at 2.85.
People’s Congress Deputies and Personal Honor and Benefits.
Whom to Side With?
The second set of questions asked the deputies to identify whom they represent: the government, the majority of the public, a particular social group, or a geographic constituency (Table 3). The set of questions was further divided into three subsets.
Whom Do Deputies Believe They Represent?
As we can see from Table 4, the first subset includes two questions: “Do you agree that deputies should help the government instead of making trouble for it (帮忙不添乱)?” and “Do you agree that deputies may challenge the government with opposing views (唱对台戏)?” Only 11.7% of the respondents strongly disagree or disagree that deputies should help the government instead of making trouble for it. By contrast, 77.8% strongly disagree or disagree that deputies may challenge the government with opposing views. At the same time, 75.9% of the respondents strongly agree or agree with the statement that deputies should not make trouble for the government, and 12.9% strongly agree or agree that deputies could challenge the government with opposing views.
Deputies Should Not Make Trouble for the Government, or Deputies May Challenge the Government.
As Table 4 shows, most deputies are relatively compliant and do not agree with challenging or making trouble for the government. However, the second subset of questions tells a different story. In Table 5, the two questions are “Do you agree that deputies should reflect citizens’ opinions when government policy is in conflict with the interests of ordinary people (老百姓)?” and “Do you agree that deputies should support government policy that is in conflict with the interests of ordinary people?” As we can see from Table 5, 92.7% strongly agree or agree with reflecting public opinion when government policy is in conflict with the interests of ordinary people. At the same time, 46% of the respondents strongly disagree or disagree with the notion that the role of deputies is to support government policy while 32.4% strongly agree or agree that deputies should support government policy.
Should Deputies Reflect Public Opinion or Support Government Policy?
Since an overwhelming majority of deputies tend to believe that they should speak up on behalf of ordinary citizens, we then asked whom deputies think they should represent. Table 6 shows the results of the third subset of questions: “Do you agree that deputies should work on behalf of the overall interests of the majority rather than the partial interests of a particular social group?” and “Do you agree that deputies should subordinate local interests to the general interest and shouldn’t represent local interests?” Only 9.7% of the respondents strongly disagree or disagree that deputies should act on behalf of the majority rather than local interests, while 5% strongly disagree or disagree that deputies should act on behalf of the public’s overall interests rather than partial interests of a particular social group.
Should Deputies Act on Behalf of Majority Interests or Partial Interests, and Majority Interests Rather Than Local Interests?
Mandate or Trustee View of Representation
The third set of questions is about the mandate or trustee view of representation. If a deputy’s personal opinion is different from public opinion, our survey asked such deputies if they would side with public opinion 民意 (mandate view) or insist on reflecting their own opinion (trustee view). The two questions we asked are “Do you agree that deputies should accurately reflect public opinion irrespective of whether they think it’s right?” and “Do you agree that deputies should reflect the opinion that they believe is right, regardless of whether it’s the same as public opinion?” Table 7 shows the results: 55.6% of the respondents strongly agree or agree that deputies should reflect public opinion no matter whether they agree with it, while 28.7% strongly disagree or disagree with this statement. At the same time, 34.8% strongly agree or agree that they should reflect their own opinion, regardless of whether it is the same as public opinion or not; 46% strongly disagree or disagree with this statement. Although more respondents believe they should reflect public opinion instead of their personal opinion, the margin is not huge.
Should Deputies Follow a Mandate or a Trustee View of Representation?
What Affects Role Perceptions?
Our study focuses on three possible independent variables to explain a deputy’s role perceptions: electoral incentives, political attitudes, and social background. We tested the correlation between the explanatory variables and the different role perceptions step by step. We also controlled for the regional setting, which is measured by the proportion of the nonagricultural population and the demographic attributes of gender, age, nationality, and education.
Electoral Incentives
We begin by exploring whether electoral incentives significantly affect deputies’ role perceptions.
We used four items to measure electoral incentives. The first is how the respondent was nominated as a county-level LPC deputy. We expected those who claimed that they were nominated by voters to be more likely to have a sense of constituency. Respondents were asked to identify how they were nominated from five options: by the parent party organization assigning them to the subordinate organization, or what is known as “being hooded” 带帽; by political parties or people’s organizations; by political parties or people’s organizations in the name of the voters; by voters jointly and independently; and “I don’t know.” If the respondent claimed he or she had been jointly and independently nominated by the voters, we coded this as a voter nomination. Conversely, if the respondent chose any of the other options, we coded it as a non-voter nomination. As Table 8 shows, 53.8% of the respondents claimed they were jointly and independently nominated by the voters.
How Deputies Reported They Were Nominated.
The high percentage of voter nominations here does not tally with the actual situation. The number of independent candidates who are jointly and independently nominated by voters is strictly controlled in China. Few people can be successfully nominated by voters; in fact, the party-state takes various measures to prevent voter nominations. Therefore, we consider the voter nominations reported by the respondents as a subjective perception of how they were nominated rather than how they were actually nominated.
If deputies hope to be reelected, one might expect that they would have an incentive to pay a great deal of attention to their constituency. We code the motivation for reelection as an ordinal variable on a 4-point scale, where 4 stands for the strongest incentive and 1 the weakest. The option of “I want to be reelected and I actively seek to be reelected” is coded as 4, “I want to be reelected but I will not actively seek to be reelected” is 3, “I don’t care whether I’m reelected or not” is 2, and “I don’t want to be reelected” is 1. As can be seen from Table 9, 49.4% of the respondents actively seek to be reelected, 42.3% want to be reelected but do not intend to actively seek reelection, 5.6% answered that they do not care whether they are reelected, and 2.8% claim they do not want to be reelected.
Deputies and the Incentive of Reelection.
Next, we asked whether the deputies hope to be promoted to a higher position in the people’s congress system, such as a deputy in a higher-level LPC deputy and/or member of the Standing Committee of the National People’s Congress. We assume that a deputy with a strong desire to be promoted would be more likely to support the government. We also coded the desire using ordinal variables on a 4-point scale, where 4 represents the strongest wish to be promoted and 1 the weakest. Table 10 shows the results: 38.0% of respondents report that they actively seek to be promoted; 46.1% reported that they hope to be promoted, but they do not actively seek to be promoted; 10.1% do not care whether they are promoted; and 5.8% do not want to be promoted.
Deputies’ Incentive to Seek Promotion.
Finally, we asked deputies about their personal judgment of the degree of competition in their election. Although once they are nominated most deputies will be elected, this does not mean potential competition is absent. We assume that deputies are more inclined to have a sense of representation when they think an election will be more competitive. Given that it is impossible for us to know the actual vote in an election, we measure the degree of competition in elections by the deputies’ subjective judgment. The respondents were asked to answer whether they agree that electoral competition is fierce. Table 11 shows the results: 61% of the respondents report that they consider electoral competition to be fierce, while 39% do not think so.
Is LPC Electoral Competition Fierce?
Political Attitudes
We take the political attitudes of deputies as an independent variable that might affect their role perceptions. This was measured in two ways. First, we looked at authoritarian values, and asked one question: “Do you agree that whether leaders are democratically elected is not important if they serve the people wholeheartedly?” The respondents were asked to answer the question by choosing from five options. Table 12 shows the results: only 25.8% of the respondents strongly disagree or disagree with the statement. By contrast, 63.1% strongly agree or agree, with 11.2% neither disagreeing nor agreeing
Deputies Authoritarian Values.
Note. The responses are to the question “Do you agree that whether leaders are democratically elected is not important if they serve the people wholeheartedly?”
The second question concerned regime support. To measure the deputies’ support for the regime, we used four items to construct an additive index. The respondents were asked to answer whether they agree with the following four statements based on a 5-point scale, where 1 stands for strongly disagree and 5 for strongly agree: “I’m satisfied with the current political institutions in China”; “In contemporary China, maintaining stability is the top priority”; “The rights of the citizens in China are well-protected”; and “The political values the government advocates are the same as mine.” To ensure internal consistency, we tested reliability in the present sample for the scale. The Cronbach α coefficient is >0.831. The internal consistency is rather good. These four items were then combined to form an additive index of a respondent’s support for the regime. Table 13 shows the results. The mean score is 17.05 for the entire index, far above the mid-point (12) of the index scale. Also, the standard deviation is rather small, which means the deputies reported that they were largely supportive of the current regime.
Deputies’ Support for the Regime.
Social Background
This article focuses on three aspects of deputies’ social background. First is CCP membership. We assume that CCP members would be more inclined to side with the government than with ordinary citizens. Table 14 presents the results: 77.7% of the respondents are CCP members.
Deputies’ Membership in the CCP.
Second is a background in state employment. We asked a multiple-choice question about whether the deputies had ever worked for the CCP, government agencies, state-owned or state-controlled enterprises, public institutions 事业单位, or collective enterprises and institutions 集体企事业单位. If the respondent chose any one of them, we coded it as a deputy with a state-employment background. As we can see from Table 15, 81.5% of the respondents have a state employment background. We also asked whether the deputy had ever been a self-employed entrepreneur or worked in a private enterprise or foreign-funded enterprise. As Table 15 shows, 21.4% of the respondents have a nonpublic-owned-enterprise employment background. A final employment background involves membership in a village or residents’ committee; 32.7% claimed that they had worked for one of these.
Deputies’ Employment Background.
Third is deputies’ administrative rank, which we coded as an ordinal variable. As we can see from Table 16, 57.9% reported they had no administrative rank. Among those with an administrative rank, section-level officials account for 21.2%, which is the largest proportion. Deputies with a deputy-section level rank and those below that level account for 5.6% and 3.5%, respectively. Deputies with a county-level or deputy county–level administrative rank account for 5.4% and 3.9%, respectively. It is noteworthy that 2.4% of the respondents were above-county-level officials.
Deputies’ Administrative Rank.
Demographic Attributes and Regional Settings
Finally, we controlled for the demographic attributes of the deputies and regional settings. In China, those who are better-educated, middle-aged, and male might have a higher level of political efficacy. It is such people who can be considered to be more likely to have a sense of representation. Nationality also needs to be controlled because a national minority county was included in the sample. In our sample, female and male deputies account for 26.3% and 73.7%, respectively. In addition, 23.4% of the respondents are ethnic minorities while 76.6% are members of the Han majority. As for age, 12.1% of the respondents were older than 55 years; 36.7% were aged between 46 and 55 years; 39.2% were 36 to 45 years old; and 11.7% were between 26 and 35 years old. The respondents younger than 25 years account for 0.3% of our sample. Regarding education, 40.6% have a bachelor’s degree or higher: 33.6% have a bachelor’s degree alone and 7% have a master’s degree or higher. At the same time, 59.4% have a college degree or lower. Of that group, 23.1% graduated from college; 22% finished high school or a technical secondary school; 13.2% only finished middle school; ten respondents, or 0.9%, reported that they only finished primary school; and one respondent reported that he never had any formal education.
We also controlled for regional settings. Recent studies have found that most independent candidates are from economically developed cities, and that such deputies are likely to be more civic-minded and to speak for ordinary people (Tang, 2004; Zou, Tang, and Huang, 2004). Since the proportion of the nonagricultural population can be used as an indicator of both economic development and urbanization, we used each county’s proportion of the nonagricultural population to measure the regional setting. Jiangyuan district in Jilin province ranks first in terms of proportion of the nonagricultural population: in Jiangyuan it accounts for 81.77% of the total population, while Debao county in Guizhou province ranks last, with just 8.76% (see Table 17).
Proportion of the Nonagricultural Population.
Results of Multilinear Regressions
In our study, we have sought the causes of significant effects on the role perceptions of county-level LPC deputies in China. First, we must address the colinearity problem. The result shows there is little colinearity between social background and political attitudes. The VIF (variance inflation factor) is less than 1.5 for all correlations. Furthermore, we added the independent variables into the model step by step. Looking at Table 18, Model 1 shows the result when the variable of electoral incentives alone was considered. Model 2 shows the result when the social backgrounds were added. Model 3 shows the result when political attitudes were included and all the explanatory variables were considered together. Finally, we added in the controlled variables, and Model 4 shows the final results when all the independent variables were combined.
Results of Multilinear Regressions.
p ≤ .1. **p ≤ .05. ***p ≤ .01.
Table 18 presents the results of the multilinear regressions. First is the perception of “personal honor and benefits,” which means that being a deputy is considered a route to personal honor and benefits. As we can see, the negative impact of a state-employment background as well as village and residents’ committee work is most significant and robust (p ≤ .01). A nonpublic-owned-enterprise employment background also has a robustly negative effect (p ≤ .05), which runs counter to the presumption that private entrepreneurs often seek to be a people’s congress deputy for social status and financial benefits. Furthermore, the desire to be promoted within the system of the people’s congresses also has a negative correlation (p ≤ .05). There are two possible explanations for this. One is that being a deputy might entail no personal benefit and thus deputies do not expect to cash in on their role. The other is that deputies may have begun to have a sense of representation and thus no longer consider serving in a people’s congress as a route to honor and personal benefits. Furthermore, the perception of electoral competition and deputies’ administrative rank also have positive effects.
Second, we asked two questions about whether the deputies were willing to challenge the government. As for the first statement, that deputies should help the government instead of making trouble, the respondent’s level of support for the regime, authoritarian values, and wish to be promoted all have significantly and robustly positive effects. The higher the regime support, the more the respondent agreed with authoritarian values. The greater the wish to be promoted, the more likely the deputy was to agree with the statement. By contrast, those who had worked in village and residents’ committees were less likely to agree with the statement (p ≤ .1). As for the second statement, that deputies can challenge the government with opposing views, those deputies with a greater desire to be promoted and those with more authoritarian values were less likely to agree with it. Furthermore, the proportion of the nonagricultural population is positively correlated with the role of challenger (p ≤ .01), which means that deputies from areas with a large urban population are more inclined to agree that they could challenge the government.
Third, we asked whom the deputies chose to side with when government policy and the interests of ordinary people conflict. It is noteworthy that a background of working for nonpublic-owned enterprises as well as in village and residents’ committees are both negatively correlated with the statement that deputies should support the government’s policy when it is in conflict with the people’s interests. The impact is significant and robust (p ≤ .01). In addition, deputies who had worked in village and residents’ committees are also more likely to reflect public opinion, while deputies with a state-employment background are less likely to support the government’s policy, but the two correlations are both of little statistical significance (p ≤ .1). By contrast, supporting the regime has a positive effect on both supporting the government’s policy and reflecting the opinions of ordinary people (p ≤ .01). That is to say, deputies who strongly support the regime tend to both reflect citizens’ opinions to the government and support government policy. For them, the two roles do not seem contradictory. Also, quite different from theoretical assumptions, voter nomination has a positive effect on agreeing with supporting government policy (p < .05). Deputies claiming that they were jointly and independently nominated by the voters are more likely to choose to support government policy. A possible explanation is that they actually had no idea how they were nominated. Among the controlled variables, the better-educated deputies tend to disagree with the statement that one should support government policy that is not consistent with the public interest (p ≤ .1).
Fourth, the deputies were asked whether they agreed that they should act on behalf of the overall public interest rather than the partial interests of a particular social group or local interests. Deputies who strongly support the regime and those with authoritarian values are more likely to agree that they should not represent solely local interests. The correlation is significant and robust (p ≤ .01). In addition, deputies with a high level of support for the regime tend to disagree with the statement that they should represent the partial interests of a particular social group (p ≤ .01). Furthermore, in cases where deputies with a high administrative rank are local leaders, administrative rank has a significant effect on whether a deputy will represent local interests. Deputies with a higher administrative rank are more supportive of local interests (p ≤ .05). The same is true of deputies with a nonpublic-owned-enterprise employment background (p ≤ .05). The voter nomination and competition perception are also positively correlated with a deputy choosing to represent overall interests instead of partial interests, although the correlation is not as significant (p ≤ .1).
Fifth, we asked whether the deputies agree with the mandate or trustee view of representation. The mandate view means deputies should accurately reflect public opinion no matter whether they think it is right, while the trustee view means they should espouse views they believe are right, no matter what public opinion may be. Deputies with a nonpublic-owned-enterprise employment background are less likely to agree with the mandate view (p ≤ .05). Those with a background in state employment are also less supportive of the mandate view, but the correlation is less significant (p ≤ .1). By contrast, deputies who had worked in village and residents’ committees tend to agree less with the trustee view (p ≤ .05). Furthermore, deputies with authoritarian values are more likely to agree with both the trustee and mandate view of representation (p ≤ .01). It seems that the potential conflict between the two views did not matter to them. In addition, deputies who are highly supportive of the regime are more inclined to agree with the mandate view; however, the impact is less significant when the controlled variables are added in (p ≤ .1). The impact of the electoral competition perception is significant in Model 4 (p ≤ .05). Deputies who perceive their election to have been relatively competitive tend to reflect public opinion even if they think it is not right, which seems to be in accordance with the assumption of electoral connections. Moreover, deputies who desire to be promoted are less inclined to agree that they should reflect their own views. Among the controlled variables, older deputies tend to agree with both the mandate view and trustee view of representation. The better-educated tend to disagree with the mandate view, while deputies from the more urbanized areas tend to support the trustee view of representation.
Discussion and Implications
Based on our national survey data, this article draws a distinction between role perceptions and role fulfillment. The results of the regressions present some interesting findings. First, it seems that county-level LPC deputies as a whole have mixed or overlapping role perceptions, which could be defined as neither state agents nor as citizen representatives. When government policy is in conflict with the interests of ordinary citizens, an overwhelming majority of the respondents chose to side with ordinary people and reflect their opinions to the government. More than half of the respondents clearly strongly disagree or disagree when asked whether they support the government, while only about 30% are definitely supportive of government policy. It seems that the deputies did speak for the people and had a sense of representation. However, in sharp contrast, most of the respondents agree that deputies should help rather than make trouble for the government and a majority of them disagree with the proposition that deputies could challenge the government with opposing views. It seems that most of deputies by no means consider themselves opponents of the state, but they insist on speaking for ordinary citizens when government policy violates the public interest. Conversely, although deputies are more inclined to side with ordinary people, this does not mean they would potentially challenge the state. When we asked whom they represented, an overwhelming majority of the respondents said they should represent the interests of the majority rather than local interests or the partial interests of a particular social group. This is rather different from findings in the recent literature, which argues that deputies tend to represent their geographic constituency or a particular social group. The difference in findings can perhaps be explained by the distinction between role perceptions and actual role fulfillment. Although the deputies with a mandate view of representation are more numerous than those with a trustee view of representation, the margin is not huge.
Second, the social background and political attitudes of county-level LPC deputies have a significant effect on their role perceptions, but electoral incentives make little difference. Deputies with a high level of support for the regime and those with authoritarian values share rather similar role perceptions: they tend to perceive their role as both a state agent and a citizen representative. They are more likely to agree that deputies should help the government rather than make trouble for it, while those with authoritarian values are definitely less inclined to agree that deputies could challenge the government. Both of them tend to agree that local interests should be subordinated to the general interest, while those with authoritarian values also tend to agree that deputies should represent partial interests. Furthermore, when government policy is not in the interest of ordinary citizens, deputies with a high level of support for the regime are inclined to both reflect the opinion of ordinary citizens and support government policy, while those with authoritarian values are more likely to agree with both the trustee and mandate views of representation.
In sharp contrast, deputies with a nonpublic-owned-enterprise employment background and those who have a background as members of village and residents’ committees share similar role perceptions and seem to be more assertive. When government policy is in conflict with the interests of ordinary people, they tend to disagree with the notion that they should support government policy. They are also less likely to consider being a deputy as a road to personal honor and personal benefits. Deputies with a nonpublic-owned-enterprise employment background also tend to disagree with the statement that deputies should represent the overall interests of the majority of the people instead of local interests and the partial interests of a particular social group. Deputies with village and residents’ committee experience are also more inclined to disagree with the trustee view of representation. In addition, a state employment background has a significant impact on several role perceptions. Deputies with such a background are less likely to consider being an LPC deputy as a means to secure personal honor and benefits. They are also less likely to agree with supporting government policy when it is in conflict with the interests of ordinary people and tend to disagree with the mandate view of representation, although the correlation is not very significant. Furthermore, those with a high administrative rank are more inclined to disagree that deputies should not represent local interests.
Electoral incentives do not seem to have the expected significant effect on most role perceptions. The incentive of reelection does not have a significant effect on any role perception when all the dependent variables are considered together. Furthermore, deputies who reported voter nomination are unexpectedly more likely to support government policy that is not inconsistent with the interests of ordinary citizens. In addition, those who desire to be promoted tend to agree that deputies should help rather than make trouble for the government. They are also more likely to disagree with the statement that being a deputy is a means to secure personal honor and benefits, and they tend to disagree with the trustee view of representation.
Regarding the controlled variables, we found that better-educated deputies are less likely to support government policy that is not in the interests of ordinary citizens. They also tend to disagree with the mandate view of representation. Deputies from more urbanized areas are more inclined to agree that they can challenge the government with opposing views and agree with the trustee view of representation. Older deputies tend to disagree that deputies should reflect public opinion when government policy is in conflict with it. They tend to agree with both the trustee view and the mandate view of representation.
What are the implications of these findings? First, the county-level LPC deputies as a whole try to strike a balance between seemingly contradictory roles, which is a distinctive characteristic of representation in China. People’s congress deputies have long been considered state agents and in the past scholars had paid insufficient attention to them. More recent studies seek to bring the deputy back in and argue that local deputies have begun to represent their constituencies both in attitude and in action. Based on our national survey, our study focuses on role perceptions rather than role fulfillment and finds that an overwhelming majority of county-level LPC deputies do perceive themselves as representatives of overall interests and even choose to side with ordinary people if government policy is not in their interest. However, they by no means challenge the current regime. Rather, their role perceptions overlap. If we emphasize any one role at the expense of the other, the distinctive character of congressional representation in China would be missed.
Second, the fact that most LPC deputies consider themselves as citizen representatives does not mean that they will gradually turn into challengers of the current regime. Recent studies reveal that deputies are increasingly responsive to their constituents or ordinary citizens. Nevertheless, our data analysis shows that most LPC deputies still stick with political correctness and are careful not to make trouble for the government. Furthermore, and unlike the latest study, which argues that the CCP is engineering a system of representation within the bounds of the National People’s Congress, fostering information revelation but silencing political activism (Truex, 2016), this article shows that bounded representation is not something that is imposed only by the CCP, but is also rooted in the overlapping role perceptions of people’s congress deputies. If we fail to recognize that role perceptions overlap, it might be inferred from China’s bounded representation that the deputies would confront the party-state to get rid of its control.
Third, congressional representation develops when deputies gradually recognize the tension between the different roles and give up their perception that roles overlap. Although electoral competition is strictly restricted in China, our survey shows that nearly all deputies stated they should act on behalf of the public interest. However, this does not herald the development of congressional representation in China and in fact might hide the differences among the deputies and the changes that are actually happening. We argue that the difference is not whether they are inclined to act on behalf of the people; rather what makes a difference is that some deputies, such as those with a high level of regime support and authoritarian values, still seek a balance between seemingly contradictory roles. Others, such as those with a nonpublic-owned-enterprise employment background, are now more sensitive to the conflicts between the roles and choose to be more assertive, although they do not want to challenge the government either. Those who realize the tensions between the roles and choose to side with their constituency are more likely, finally, to develop a sense of representation, even though they might still be restricted by the political structure and institutions from truly acting as citizen representatives.
Footnotes
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 disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors would like to acknowledge the support received from the National Social Science Fund (No. 15BZZ008), Fundamental Research Funds for the Central Universities (No. 17WKZD23), MOE Liberal Arts and Social Sciences Foundation (No. 14YJC810005), as well as the Youth Pearl River Scholarship.