Bread or Roses: How Economic Inequality Affects Regime Support in China?

Abstract

Much comparative politics scholarship has examined whether economic inequality affects democratic values or political support in democracies. Nevertheless, they lack a close examination of the political effects of economic inequality and, more importantly, how economic inequality shapes political support in non-democracies. We provide an empirical test of the effect of economic inequality on regime support using the China data from the Asian Barometer Survey between 2002 and 2015. We argue and demonstrate that perceived economic inequality significantly reduces regime support in China. Moreover, using a causal mediation analysis, we find that the detrimental effect of perceived economic inequality on regime support is not driven by demands for redistribution, but rather by the political value orientation. These findings advance our understanding of the connection between economic inequality and political values and the economic base of political legitimacy in non-democracies.

Keywords

Asian Barometer Survey authoritarian values China economic inequality redistribution regime support

Rising economic inequality has become one of the most challenging issues in the world. Since the turn of the century, much comparative politics scholarship has examined the effect of economic inequality on redistributive preferences (Finseraas, 2009), democratic values (Bratton and Mattes, 2001; Krieckhaus et al., 2013), and political support (Anderson and Singer, 2008; Perrella et al., 2016) in democracies. By contrast, empirical research that explicitly connects economic inequality to regime support in non-democracies is scant. The political economy literature on democratization posits that economic inequality incentivizes the poor to revolt with the expectation that democracies are more likely to redistribute income from the rich (Acemoglu and Robinson, 2006; Boix, 2003). Some recent studies, however, demonstrate that the link between economic inequality and democratization is less tenable than predicted (Ansell and Samuels, 2014; Haggard and Kaufman, 2012). Does economic inequality affect regime support in authoritarian regimes? How does economic inequality shape individuals’ attitudes toward the current regime in the authoritarian setting? These questions relate to two general inquiries in comparative politics: the economic base of political support and the legitimacy of authoritarian regimes.

In theory, the effect of economic inequality on political support is strong, if not stronger, in the authoritarian setting for several reasons. First, citizens in authoritarian regimes usually pay more attention to economic outcomes than to procedures in evaluating the regime (Ou-Yang and Zhou 2019). Second, authoritarian rule features a blur between political regime and government as well as heavy governmental involvement in the economy, under which citizens are more likely to attribute the responsibility of negative economic outcomes—including uneven development—to the current regime (Huhe and Tang, 2017; Tang et al., 2015). Third, the resentment over inequality can be periodically vented by election under democracies; without this venting mechanism, citizens’ perception of inequality may lead to the preference for a change in the political system under a high level of inequality (Chu and Wu, 2020). The lack of attention toward the relationship between economic inequality and regime support in authoritarian regimes is therefore curious. Moreover, it remains unclear exactly how economic inequality can shape political support in authoritarian countries.

We attempt to fill the gap in the literature by analyzing the effect of economic inequality on regime support and exploring the mechanisms through which economic inequality shapes regime support. We posit that China is a crucial case for this study because Chinese are found tolerant of inequality (Wu, 2009), and their trust toward the current political regime remains high (Chen, 2017; Lu and Dickson, 2020) despite the soaring inequality attributable to state policies.¹ Some argue that continuous and phenomenal economic growth offsets the adverse effect of economic inequality on political support and that citizens’ regime support is determined more by government economic performance than distributive outcomes (Chi and Kwon, 2016). In this sense, China is an important case to probe the trade-off between economic growth and economic inequality in the formation of regime support. If economic inequality is found exerting a negative effect on regime support in China, the support-eroding effect of economic inequality should be even more pronounced in authoritarian countries with poorer economic growth records.

China’s socialist legacy and ruling ideology from the Maoist era make economic inequality a prominent threat to its regime legitimacy. Nonetheless, research on inequality in post-Mao China has yielded mixed findings regarding its political consequences. Some sociological research finds that most Chinese consider existing levels of inequality acceptable, and that economic inequality is not the main source of public discontent directed at the Chinese government (Whyte, 2010; Wu, 2009). By contrast, recent political studies have revealed the detrimental effects of both objective and perceived inequality on political support in China (Chu and Wu, 2020; Zhou and Jin, 2018). Behind these disparate findings lies a theoretical debate about the political effects of economic inequality. On the one hand, political economy literature on democratization posits that economic inequality increases public demands for redistribution and, thus, subverts regime support and induces regime change in autocracies. On the other hand, demands for redistribution may strengthen public support for the authoritarian regime as both the rich and the poor would be supportive of the authoritarian government’s role in redistribution (Lai et al., 2021; Park, 2020). We add new insights to this theoretical debate by examining the effect of economic inequality on regime support and pinpointing the mechanism that connects economic inequality to regime support in China.

Using data collected from mainland China in four waves of the Asian Barometer Survey (ABS) between 2002 and 2015, we demonstrate that the perception of economic inequality significantly dampens popular support for the current political regime in China. Employing a causal mediation analysis, we further find that the support-eroding effect of perceived economic inequality is not driven by demands for redistribution; rather, political value orientation plays an important mediating role between perceived economic inequality and regime support. Specifically, around 38% of the negative effect of perceived economic inequality on regime support is attributed to the weaker authoritarian value orientation under inequality.

Our study differs from the existing studies on inequality and political support in two ways. First, whereas existing studies focus on whether inequality affects political attitudes (Chu and Wu, 2020; Park, 2020; Zhou and Jin, 2018), we put our emphasis on the mechanisms through which economic inequality shapes regime support. Second, whereas most existing studies in the vein of modernization theory have examined the effect of economic development on values and attitudes (Wang, 2005; Welzel and Inglehart, 2009), we pay special attention to whether and how economic distribution leads to changes in political values and attitudes that sustain regime legitimacy and stability. Thus, this article makes the following contributions.

First, we link the political importance of economic equality in authoritarian settings with the operation of two mechanisms: redistributive demands and political value orientations. Adding to the research that casts doubts on redistribution-driven democratization in autocracies (Ansell and Samuels, 2014; Haggard and Kaufman, 2012), our study shows that in China, perceived economic inequality is not correlated with demands for redistribution and that this null effect holds for both the rich and the poor. Instead, the detrimental effect of perceived economic inequality on regime support works primarily through reducing individuals’ affective identification with authoritarian values such as unquestioning respect for authority—the intrinsic source of regime support in autocratic countries. This sheds a very different light on why economic inequality can be politically destabilizing.

Second, while instrumental (or material) interests and intrinsic values are often treated as separate sources of regime support (Chu, 2013; Lu and Dickson, 2020), our study demonstrates the crucial interaction of these two in the formation of regime support. Modernization theory posits that economic development fosters emancipative values that give “priority to equality over patriarchy, tolerance over conformity, autonomy over authority, and expression over security” (Welzel and Inglehart, 2009: 129; Lipset, 1960). In line with this theory, we show that economic distribution also brings about value change. In a country with lingering Confucian traditions like China, an individual’s affective identification with authoritarian values, such as deference to authority, group primacy, and paternalism, is not immutable. In fact, authoritarian value orientation is subject to both macro-level income inequality and micro-level perception of economic inequality. Broadly, our findings are consistent with modernization theory, providing evidence for the change in political values resulting from socioeconomic change.

Third, our findings have important implications for the legitimacy and stability of Chinese authoritarian regime. We find that the perceived economic inequality significantly dampens regime support. This confirms the existence of the “social volcano” in China—negative effects or views of economic inequality among citizens (Lü, 2014a; Zhou and Jin, 2018). Moreover, we find that the enhancement of regime support brought about by perceived economic growth outweighs the pernicious effect of perceived economic inequality on regime support. This explains why the Chinese support for the current regime remains strong despite the growth of economic inequality in recent decades. To the extent that China’s economic growth is slowing down and the inequitable distribution of wealth is not effectively corrected, our study reveals the fragility of the growth-based legitimacy of Chinese authoritarian regime.

Theoretical Expectations

Economic voting literature has found that government’s economic performance is an important predictor of political support in both democracies and non-democracies (Lewis-Beck and Stegmaier, 2000; Rose and Mishler, 2002). The expansion of economic inequality indicates a lack of will or capacity in the current political system to rectify inegalitarian outcomes that are associated with other problems, including high crime rates and corruption, which can lower popular support toward the regime. In particular, the connection between economic outcomes, including the distribution of wealth, and regime support is strong in China for several reasons.

First, autocracies need to generate satisfactory economic performance and ensure good governance to compensate for the deficits in procedure legitimacy (Beloshitzkaya, 2020; Gilley, 2006). This effect reflects the performance legitimacy thesis in China, where the Chinese Communist Party (CCP) garners public support through economic growth (Chen, 2017; Lu and Dickson, 2020) and expansion of social welfare provisions (Huang and Gao, 2017; Lü, 2014b).

Second, different from the complex institutional designs in democracies for power sharing that blur the “clarity of responsibility” and make it difficult for voters to assign credit and blame to the government (Anderson, 2007), power in authoritarian regimes is concentrated in the hands of a specific leader or institute, facilitating the interpretation of economic outcomes as government responsibility. The Chinese government has great control over the economy and frequently intervenes in the economy through asset ownership and investment (Huhe and Tang, 2017). Consequently, Chinese are more likely to credit or blame the government, which fuses with the political regime, for economic outcomes.

Third, the Chinese authoritarian regime embraces economic performance in its propaganda and legitimation. Like nondemocratic governments in South Korea and various Latin American countries in the 1960s and 1970s (Diamond et al., 1988), economic prosperity lies at the center of the rose-tinted future that the CCP promised in post-Mao China. As such, individuals are indoctrinated to base their evaluation of the political system on its economic performance or outcomes.

Therefore, we formulate a first-stage hypothesis about the effect of economic inequality on regime support as follows:

Hypothesis 1. Economic inequality decreases regime support in China.

We further examine the mechanisms that may drive the relationship of economic inequality and regime support in China. We draw on existing studies of political support that have identified two sources of regime support—material interest and value orientation. The former postulates that citizens’ political support is a function of the material benefits that the government provides to them (Riker, 1990). The latter suggests that regime support stems not just from the functioning of the political system, but also from the prevailing political predispositions held by its citizenry. As such, the effect of economic inequality on regime support might operate through individuals’ calculations of material well-being and/or individuals’ value orientations. In what follows, we propose a hypothesis for each mechanism.

The median voter theorem (Meltzer and Richard, 1981) posits that democracy is closely associated with redistribution because the majority of the population is poor and would vote for higher levels of tax-cum-transfer or income redistribution from the rich. In this circumstance, democratic institutions function as means through which elites can credibly commit to the continuation of redistributive policies over time. Hence, high inequality will increase support for democratization among the masses through enhancing their demands for redistribution (Acemoglu and Robinson, 2006; Boix, 2003). Accordingly, one might expect that economic inequality decreases regime support in autocracies by increasing the masses’ demands for redistribution.

However, recent studies have cast doubts on such predictions about the relations of economic inequality, redistributive preference, and regime support. In advanced capitalist economies, inequality might not enhance demands for redistribution because redistributive preferences are mediated by many other factors, such as beliefs about the fairness of the distribution (Alesina et al., 2012), prospect of upward mobility (Benabou and Ok, 2001), and trust in government (Hetherington, 2005). A growing body of literature on social welfare preferences in developing and post-communist countries suggests that in unequal societies, rather than stronger, the poor’s demand for social welfare is actually weaker because they did not previously receive such benefits (Haggard et al., 2013; Pop- Eleches and Tucker, 2017). Previous research in China also finds that economic inequality does not lead to a strong demand for systematic redistribution because most respondents considered economic inequality as merit-based and fair (Whyte, 2010). Moreover, in communist and post-communist countries, the public strongly expects the government to provide social welfare (Beloshitzkaya, 2020; Kuhn, 2013; Pop- Eleches and Tucker, 2017). Such expectations might strengthen, rather than weaken, support for the authoritarian regime as democracies do not necessarily redistribute more (Beloshitzkaya, 2020; Knutsen and Wegmann, 2016).

Given the contradictory predictions regarding the relations of economic inequality, redistributive preference, and regime support in non-democracies, we derive a hypothesis about the redistribution mechanism through which economic inequality decreases regime support in China, without specifying the exact signs of coefficients in the links between economic inequality and redistributive demand and between redistributive demand and regime support:

Hypothesis 2. Economic inequality decreases regime support in China through demands for redistribution.

The second mechanism through which economic inequality decreases regime support is built on the literature about how economic inequality influences individuals’ identification with the political values underpinning the regime. Economic inequality, especially when a lack of equal opportunity and the inaction of government result in a decreasing share of net economic benefits for the poor, intensifies the feeling that society is breaking down. This will invoke a crisis of the core values underpinning the political regime. Cross-national studies in democracies have found that growing income gaps have adverse impacts on identification with democratic values (Anderson and Singer, 2008; Krieckhaus et al., 2013). Similarly, economic inequality in authoritarian settings generates political disillusionment and reduces individuals’ identification with political values that sustains authoritarianism. Specifically, we argue that economic inequality weakens individuals’ affections for authoritarian values—preferences for concentration of power, conformity and order, and hierarchy, respectively.

First, economic inequality is closely associated with political inequality (Houle, 2018; Solt, 2008). In highly unequal societies, ordinary people tend to link economic inequality to corruption and a manipulative distribution of power and resources that leads to greater inequality (Ariely and Uslaner, 2017). Existing research shows that the perception of inequality leads to stronger social expectation for anti-corruption reform in China (Li et al., 2016). It is equally plausible that economic inequality fosters a greater desire for power sharing that can reduce corruption and political inequality.

Second, the masses are motivated to change the status quo in times of economic inequality. Extant studies find that support for left-wing political orientations tends to rise in highly unequal societies with legacy of an authoritarian regime (Visser et al., 2014). This is because individuals under such a circumstance are easily motivated and mobilized by left-wing ideologies to challenge the established order that is deemed unfair and ineffective. Moreover, individuals on the political left attribute less importance to power, order, and conformity (Jost and Hunyady, 2002). Therefore, support for authoritarian values, such as appreciation of order and conformity, may decline in a highly unequal society.

Third, people’s intrinsic aversion toward economic inequality leads to disapproval of hierarchical power relations. The emphasis on hierarchical social and political relationships in authoritarian values is antithetical to the idea of equality. Empirical evidence shows that a significant proportion of population are inherently averse to inequality and prefer lower aggregate levels of income gaps (Lü and Scheve, 2016; Osberg and Smeeding, 2006). Those self-reporting that “economic inequality situation in my country is very high” are more likely to resent inequality. This population’s aversion to inequality makes them less likely to support the authoritarian values that uphold hierarchies, which are essentially unequal social and political relationships.

Regarding the link between political value orientation and regime support, it has been established that individuals evaluate the political system’s performance in ways that are consistent with their political values, particularly their views toward political authority (Anderson and Singer, 2008; Bratton and Mattes, 2001). Specifically, cultural values congruent with democratic principles promote the development of a democratic regime with widespread popular support, whereas authoritarian values lead to support for a nondemocratic regime (Chu, 2013; Rose et al., 2011). As such, the declining identification with authoritarian values erodes regime support in authoritarian settings. We, therefore, formulate a hypothesis about the political value mechanism for the relationship of economic inequality and regime support:

Hypothesis 3. Economic inequality decreases regime support in China by eroding authoritarian value orientation that sustains support for the current regime.

In summary, we first predict that economic inequality has a negative effect on regime support. This negative effect can be further parsed into direct effect and indirect effect (Figure 1). The indirect effect, or mediation effect, is predicted to run through two intermediary factors (mediators): redistributive preference and authoritarian value orientation.

Figure 1.

Effects of Economic Inequality on Regime Support.

Data and Empirical Analysis

To test the hypotheses, we used the China data from four waves of ABS collected between 2002 and 2015. The survey uses probability proportional to population design, and the sample covers all provincial units of mainland China except Tibet.² There are 3183, 5098, 3473, and 4068 valid respondents in the waves of 2002, 2007, 2011, and 2015, respectively, with the total being 15,822. Most ABS survey questions, especially those related to political values and attitudes, are consistent across waves. Hence, ABS provides valuable cross-sectional data at the individual level to study Chinese political attitudes over the past decade.³

Dependent Variable

The dependent variable is an index of support for the current political regime. We measured respondents’ regime support based on the extent of their agreement with the following statements:

Over the long run, our system of government is capable of solving the problems our country faces;

Thinking in general, I am proud of our system of government;

A system like ours, even if it runs into problems, deserves the people’s support;

I would rather live under our system of government than any other that I can think of.

Answers to these questions range from 1 (very much agree) to 4 (very much disagree), and we recoded the answers so that higher values correspond to a higher level of regime support. We constructed the index of regime support by taking an average of valid answers to these four questions and using it as the dependent variable, support. We conducted factor analysis and Mokken scale analysis (reported in Tables A1–A3 in Online Appendix) to ensure the validity of this construct.

Notably, regime support is conceptually distinct and operationalized separately from individuals’ authoritarian values. First, authoritarian values are political norms that are relatively more persistent, while regime support is a political attitude or opinion that is more subject to change (Shi, 2015). Second, the questions used in constructing the regime support variable do not specify the nature of the regime or political system being supported. According to Chu (2013), Chinese respondents do not tend to formulate and report their support for the regime based on whether the regime or government is democratic or not. Third, the construction of authoritarian value orientation, to be discussed later, does not include preferences for regime type or attitudes toward the current regime.⁴

Independent Variables

The independent variable in this study, economic inequality, is captured by both objective and subjective measures. Extant studies find that subjective inequality can be quite different from objective inequality (Bavetta et al., 2020; Gimpelson and Treisman, 2018). It is the perceived, rather than actual, level of inequality that is strongly correlated with the demand for redistribution and the existence of political contention (Gimpelson and Treisman, 2018; Kuhn, 2013). We measure both subjective and objective inequality to see which one matters more for regime support in the Chinese context.⁵ Subjective inequality is based on answers to the survey question “What do you think of the economic inequality situation in our country?,” ranging from 1 (extremely low) to 5 (extremely high). For the sake of simplicity in interpreting the causal mediation analysis results, we created a binary variable, perceived inequality, as a measure of subjective inequality, with value “1” corresponding to 4–5 on the 5-point scale of the answers and value “0” to 1–3 on the 5-point scale.⁶ The objective level of inequality is measured by the Gini coefficient at the provincial level, computed by Weimin Tian (2012) based on household income data. Being consistent with the subjective measure of inequality, we convert the Gini coefficient into a binary variable, income inequality, with values above the median coded as “1” (high) and values equal to or below median as “0” (low).

Mediator Variables

The first mediator variable, redistributive preference, is constructed based on respondents’ answers to a commonly used survey question to capture preference for redistribution (Haggard et al., 2013; Kuhn, 2013; Lai et al., 2021) “Do you agree or disagree with the statement: it is the responsibility of the government to reduce the differences between people with high income and those with low income.” We code “agree” or “completely agree” with the statement as value 1, and “disagree” and “completely disagree” as value 0 for the binary variable, redistributive preference.⁷

To measure the second mediator variable, authoritarian values, we started with the most inclusive measure suggested by the literature using a battery of questions (listed in Table 1) that broadly tap into individuals’ authoritarian orientations, including their attitudes toward concentration of power (items 6, 7, 8, 10), preference for hierarchy (items 1, 2, 3), and preference for conformity and order (items 4, 5, 9). These questions (except item 2) pose statements that conform with authoritarianism, so agreement with them signals that the respondent adheres to or identifies with authoritarian values. Answers to these questions are on the 4-point scale, with value 1 meaning “very much agree” and value 4 meaning “very much disagree.” To compare the magnitude of effects, we constructed the variable of authoritarian value orientation in the same scale of redistributive preference. Specifically, we created a binary indicator for each of these questions, with value 1 corresponding to “agree” or “very much agree” and value 0 to “disagree” or “very much disagree”;⁸ then we averaged these indicators for each respondent to get an index, authoritarianism, that captured the collective profile of the respondent’s authoritarian values. The index ranges from 0 to 1, with higher values indicating higher authoritarian orientation.

Table 1.

ABS Survey Questions on Authoritarian Values.

Preference	Item #	Statement
Hierarchy	1	Women should not be involved in politics as much as men.
	2	People with little or no education should have as much say in politics as highly educated people.
	3	Government leaders are like the head of a family; we should all follow their decisions.
Conformity and order	4	The government should decide whether certain ideas should be allowed to be discussed in society.
	5	Harmony of the community will be disrupted if people organize lots of groups
	9	If people have too many ways of thinking, society will be chaotic.
Concentration of power	6	When judges decide important cases, they should accept the view of the executive branch.
	7	If the government is constantly checked by the legislature, it cannot possibly accomplish great things.
	8	If we have political leaders who are morally upright, we can let them decide everything.
	10	When the country is facing a difficult situation, it is ok for the government to disregard the law in order to deal with the situation.

Control Variables

We included additional variables to control for individuals’ demographic, socioeconomic, and sociopolitical characteristics, including age, gender, marital status, ethnicity, religion, education, per capita household income, hukou status, employment status, social status, social trust, political interest, media use for political information, and CCP membership. Moreover, we included a variable about individuals’ perception of national economic situation⁹ because economic evaluation is a strong determinant of regime support in China (Lu and Dickson, 2020; Ou-Yang and Zhou 2019). Furthermore, we controlled for individuals’ perceived prospects of upward mobility¹⁰ that are found to moderate the attitudinal effect of economic inequality in other studies (Alesina and La Ferrara, 2005; Houle and Miller, 2019). In addition, as some scholars find that perception of corruption significantly weakens regime support in China (Zhu et al., 2019), we added individuals’ perception of corruption as a control variable. For unobserved regional conditions, we included an indicator of urban/rural area as well as province-fixed effects. We also controlled for unobservable year-specific factors by including survey-year fixed effects. The definition and summary statistics of all variables are provided in Table A5 in Online Appendix.

Estimation Method

Conventional regression models can identify the average total effect of treatment (ATE) but cannot directly identify the average indirect or average mediation effect (AME) and average direct effect (ADE) because the latter two imply a counterfactual that cannot be observed in real-world settings. The analytical focus of this study is not only the correlation between economic inequality and regime support, but also the mechanisms through which economic inequality shapes regime support. The causal mediation analysis (Imai et al., 2011; Tingley et al., 2014) is, therefore, an appropriate method for this study. Unlike the conventional methods of mediation analysis (e.g. linear structural equation model), the causal mediation analysis requires weaker and more attainable assumptions¹¹ and allows for more flexible modeling. Specifically, the causal mediation analysis computes the ATE that is averaged over all respondents in the survey as in conventional methods. Moreover, it quantifies the AME that reveals the magnitude of the causal effect of economic inequality on regime support, which can be attributed to the change in redistributive preference and authoritarian value orientation, respectively, because of economic inequality. It also quantifies the ADE that measures the causal effect of economic inequality on regime support that can be attributed to causal mechanisms other than the specified indirect effect acting through the hypothesized mediators.

The causal mediation analysis was conducted in two stages: estimation and simulation. In the estimation stage, the analysis fits regressions with the mediator variables as the dependent variables (i.e. the mediator model) and then fits regressions with the outcome of interest as the dependent variable (i.e. the outcome model). Both mediator and outcome models control for relevant covariates. Based on the regression parameters estimated, the causal mediation analysis proceeds to simulate the predicted values of unobserved mediating and outcome variables from the observed variables based on quasi-Bayesian Monte Carlo approximation and then computes the AME, ADE, and ATE with confidence intervals. We estimated the mediator and the outcome models using multivariate linear regressions. For comparison, we also estimated the outcome model excluding the mediators (i.e. the baseline model). The problems concerning missing data are addressed via multiple imputations to avoid generating inefficient or biased results.

Empirical Results

Table 2 reports the regression results for subjective inequality (i.e. perceived inequality), and Table 3 reports the results for objective inequality (i.e. income inequality). Table 4 reports the results of causal mediation analysis, including AME, ADE, ATE, and the proportions of the total effect mediated by redistributive preference and authoritarian value orientation, respectively.

Table 2.

OLS Regression Results of Mediator and Outcome Models.

	Mediator model		Outcome model		Baseline
	1	2	1	2	1
	DV: redistributive preference	DV: authoritarian values	DV: regime support		DV: regime support
Perceived inequality	−0.0034 (0.0074)	−0.0416*** (0.0039)	−0.0236** (0.0072)	−0.0146* (0.0071)	−0.0236** (0.0072)
Redistributive preference			0.0089 (0.0077)
Authoritarian values				0.2165*** (0.0144)
Age	0.0015*** (0.0002)	0.0029*** (0.0001)	0.0035*** (0.0002)	0.0029*** (0.0002)	0.0036*** (0.0002)
Male	−0.0120+ (0.0064)	−0.0189*** (0.0034)	−0.0098 (0.0062)	−0.0058 (0.0062)	−0.0099 (0.0062)
Married	−0.0131 (0.0087)	−0.0101* (0.0046)	0.0138 (0.0084)	0.0159+ (0.0084)	0.0137 (0.0084)
Ethnic minority	0.0067 (0.0124)	0.0226*** (0.0066)	0.0182 (0.0120)	0.0133 (0.0120)	0.0182 (0.0120)
Religion	0.0049 (0.0094)	−0.0062 (0.0050)	−0.0076 (0.0091)	−0.0062 (0.0091)	−0.0075 (0.0091)
Middle school	0.0224** (0.0084)	−0.0182*** (0.0045)	−0.0025 (0.0082)	0.0016 (0.0081)	−0.0023 (0.0082)
High school	0.0222* (0.0094)	−0.0581*** (0.0050)	−0.0282** (0.0091)	−0.0154+ (0.0091)	−.0280** (.0091)
College	−0.0525*** (0.0157)	−0.0418*** (0.0084)	−0.0265+ (0.0153)	−0.0179 (0.0152)	−0.0270+ (0.0153)
Urban hukou	0.0360*** (0.0099)	−0.0355*** (0.0053)	−0.0308** (0.0096)	−0.0228* (0.0096)	−0.0304** (0.0096)
Income	−0.0100*** (0.0021)	−0.0144*** (0.0011)	−0.0087*** (0.0020)	−0.0057** (0.0020)	−0.0088*** (0.0020)
Farmer	−0.0151+ (0.0084)	0.0211*** (0.0045)	0.0479*** (0.0082)	0.0432*** (0.0081)	0.0478*** (0.0082)
Social status	−0.0031 (0.0035)	0.0021 (0.0019)	0.0181*** (0.0034)	0.0176*** (0.0034)	0.0181*** (0.0034)
Employed	−0.0184* (0.0088)	0.0033 (0.0047)	−0.0195* (0.0086)	−0.0203* (0.0085)	−0.0196* (0.0086)
CCP member	−0.0082 (0.0107)	−0.0323*** (0.0057)	0.0258* (0.0104)	0.0327** (0.0103)	0.0258* (0.0104)
Trust	0.0023 (0.0064)	−0.0243*** (0.0034)	0.0608*** (0.0062)	0.0661*** (0.0061)	0.0608*** (0.0062)
Political interest	0.0142* (0.0066)	−0.0145*** (0.0035)	0.0151* (0.0064)	0.0184** (0.0064)	0.0152* (0.0064)
New media	−0.0107 (.0106)	−0.0213*** (0.0057)	0.0054 (0.0103)	0.0099 (0.0102)	0.0053 (0.0103)
Urban area	0.0020 (0.0089)	0.0044 (0.0048)	0.0011 (0.0087)	0.0001 (0.0086)	0.0011 (0.0087)
Social mobility	−0.0057 (0.0040)	0.0098*** (0.0022)	0.0289*** (0.0039)	0.0267*** (0.0039)	0.0289*** (0.0039)
Perceived economic development	0.0049 (0.0078)	0.0072+ (0.0041)	0.1261*** (0.0075)	0.1246*** (0.0075)	0.1262*** (0.0075)
Perceived corruption	−0.0023 (0.0030)	−0.0126*** (0.0016)	−0.0223*** (0.0030)	−0.0202*** (0.0029)	−0.0229*** (0.0030)
Intercept	0.7014*** (0.0831)	0.6414*** (0.0444)	2.7893*** (0.0810)	2.6566*** (0.0808)	2.7243*** (0.0804)
Year fixed effects	Yes	Yes	Yes	Yes	Yes
Province fixed effects	Yes	Yes	Yes	Yes	Yes
Adjusted R²	0.1889	0.2003	0.1457	0.1577	0.1457
No. of observations	15,822	15,822	15,822	15,822	15,822

CCP: Chinese Communist Party; DV: dependent variable; OLS: ordinary least squares.

***

0.001; **0.01; *0.05; +0.10.

Table 3.

OLS Regression Results of Mediator and Outcome Models.

	Mediator model		Outcome model		Baseline
	1	2	1	2	1
	DV: redistributive Preference	DV: authoritarian values	DV: regime support		DV: regime support
Income inequality	0.0022 (0.0064)	−0.0063+ (0.0034)	0.0031 (0.0063)	0.0045 (0.0062)	0.0031 (0.0063)
Redistributive preference			0.0075 (0.0077)
Authoritarian values				0.2109*** (0.0143)
Age	0.0015*** (0.0002)	0.0029*** (0.0001)	0.0035*** (0.0002)	0.0029*** (0.0002)	0.0035*** (0.0002)
Male	−0.0102 (0.0064)	−0.0195*** (0.0034)	−0.0093 (0.0062)	−0.0053 (0.0062)	−0.0094 (0.0062)
Married	−0.0110 (0.0087)	−0.0089+ (0.0046)	0.0148+ (0.0084)	0.0166* (0.0084)	0.0147+ (0.0084)
Ethnic minority	0.0035 (0.0116)	0.0193** (0.0062)	0.0423*** (0.0112)	0.0382*** (0.0112)	0.0423*** (0.0113)
Religion	0.0065 (0.0090)	−0.0069 (0.0048)	−0.0135 (0.0087)	−0.0120 (0.0087)	−0.0135 (0.0087)
Middle school	0.0215* (0.0084)	−0.0181*** (0.0045)	−0.0043 (0.0081)	−0.0003 (0.0081)	−0.0041 (0.0081)
High school	0.0222* (0.0094)	−0.0575*** (0.0050)	−0.0287** (0.0091)	−0.0164+ (0.0091)	−0.0286** (0.0091)
College	−0.0558*** (0.0157)	−0.0433*** (0.0084)	−0.0257+ (0.0153)	−0.0170 (0.0152)	−0.0262+ (0.0153)
Urban hukou	0.0369*** (0.0098)	−0.0355*** (0.0052)	−0.0266** (0.0095)	−0.0189* (0.0095)	−0.0264** (0.0095)
Income	−0.0111*** (0.0021)	−0.0152*** (0.0011)	−0.0093*** (0.0020)	−0.0062** (0.0020)	−0.0094*** (0.0020)
Farmer	−0.0108 (0.0083)	0.0224*** (0.0045)	0.0509*** (0.0081)	0.0461*** (0.0081)	0.0508*** (0.0081)
Social status	−0.0027 (0.0035)	0.0035+ (0.0019)	0.0158*** (0.0034)	0.0150*** (0.0034)	0.0158*** (0.0034)
Employed	−0.0183* (0.0088)	0.0043 (0.0047)	−0.0205* (0.0086)	−0.0215* (0.0085)	−0.0206* (0.0086)
CCP member	−0.0071 (0.0107)	−0.0331*** (0.0057)	0.0271** (0.0104)	0.0340** (0.0103)	0.0270** (0.0104)
Trust	0.0006 (0.0063)	−0.0246*** (0.0034)	0.0608*** (0.0062)	0.0660*** (0.0061)	0.0608*** (0.0062)
Political Interest	0.0153* (0.0066)	−0.0153*** (0.0035)	0.0182** (0.0064)	0.0215*** (0.0064)	0.0183** (0.0064)
New media	−0.0117 (0.0106)	−0.0206*** (0.0057)	0.0028 (0.0103)	0.0071 (0.0102)	0.0027 (0.0103)
Urban area	0.0046 (0.0088)	0.0032 (0.0047)	−0.0042 (0.0086)	0.0035 (0.0085)	0.0042 (0.0086)
Social mobility	−0.0061 (0.0041)	0.0104*** (0.0022)	0.0282*** (0.0039)	0.0259*** (0.0039)	0.0281*** (0.0039)
Perceived economic development	0.0069 (0.0077)	0.0085* (0.0041)	0.1294*** (0.0075)	0.1277*** (0.0075)	0.1295*** (0.0075)
Perceived corruption	−0.0025 (0.0030)	−0.0120*** (0.0016)	−0.0228*** (0.0030)	−0.0203*** (0.0029)	−0.0229*** (0.0030)
Perceived inequality	−0.0049 (0.0073)	−0.0440*** (0.0039)	−0.0263*** (0.0072)	−0.0171* (0.0071)	−0.0263*** (0.0072)
Intercept	0.7124*** (0.0840)	0.5750*** (0.0166)	2.6661*** (.0308)	2.5511*** (0.0311)	2.6724*** (0.0301)
Year fixed variables	Yes	Yes	Yes	Yes	Yes
Province fixed effects	No	No	No	No	No
Adjusted R²	0.1828	0.1924	0.1382	0.1497	0.1382
No. of observations	15,822	15,822	15,822	15,822	15,822

CCP: Chinese Communist Party; DV: dependent variable; OLS: ordinary least squares.

***

0.001; **0.01; *0.05; +0.10.

Table 4.

Results of Causal Mediation Analysis.

Treatment	Mediator	Effect	95% confidence interval
Treatment	Mediator	Effect	Estimate	Loss	Upper	p value
Perceived inequality	Redistributive preference	AME	−0.0000	−0.0003	0.00	0.70
		ADE	−0.0236	−0.0370	−0.01	0.00***
		ATE	−0.0237	−0.0370	−0.01	0.00***
		Prop. mediated	0.0007	−0.0066	0.01	0.70
	Authoritarian values	AME	−0.0090	−0.0110	−0.01	0.00***
		ADE	−0.0145	−0.0283	0.00	0.00***
		ATE	−0.0235	−0.0375	−0.01	0.00***
		Prop. mediated	0.3813	0.2252	1.09	0.00***
Income inequality	Redistributive preference	AME	0.0000	−0.0001	0.00	0.81
		ADE	0.0032	−0.0101	0.02	0.62
		ATE	0.0033	−0.0101	0.02	0.62
		Prop. Mediated	0.0004	−0.1070	0.09	0.93
	Authoritarian values	AME	−0.0014	−0.0028	0.00	0.056+
		ADE	0.0043	−0.0082	0.02	0.498
		ATE	0.0030	−0.0096	0.01	0.646
		Prop. mediated	−0.1130	−2.6675	3.00	0.678

AME: average mediation effect; ADE: average direct effect; ATE: average total effect of treatment.

***

0.001; +0.10.

Effects of Economic Inequality

When economic inequality is measured by perception, the outcome-model results in Table 2 indicate that economic inequality is negatively associated with regime support. This confirms Hypothesis 1 about the total effect of economic inequality on regime support. Specifically, the regime support by individuals with high perceived inequality is lower than the one by individuals with low perceived inequality by around 0.02 point on the 4-point scale. It is worth noting that the effect of perceived inequality on regime support does not significantly differ with respondents’ income or social status.¹² Moreover, the detrimental effect of perceived economic inequality is about one-fifth of the enhancement effect of perceived economic development on regime support.

The regression results in Table 2 also show that perceived economic inequality has no statistically significant effect on redistributive preference, nor is the redistributive preference significantly associated with regime support. Hypothesis 2 is thus rejected. In contrast, perceived inequality is significantly and negatively associated with authoritarian value orientation, and authoritarian value orientation is significantly and positively associated with regime support. Specifically, respondents with high perceived inequality tend to have lower authoritarian value orientation by approximately 0.04 point on the 0–1 scale. Meanwhile, the lower authoritarian value orientation significantly decreases the respondents’ regime support by as much as 0.22 point on the 4-point scale. Comparing the results of the outcome model and the baseline model (i.e. the outcome model excluding the mediators), the coefficient of perceived inequality declines by approximately 38%, while the R² of the outcome model increases by approximately 8% when the authoritarianism variable is included, indicating a substantial mediating effect of authoritarian value orientation. All the links expected in Hypothesis 3, from economic inequality to regime support running through authoritarian value orientation, are corroborated when economic inequality is measured by perception.

Table 3 shows that the objective level of economic inequality, measured by the provincial-level Gini index, is negatively associated with authoritarian value orientation, which runs contrary to Solt’s finding from a cross-national study that individuals are more likely to view hierarchical relations as natural and hold greater respect for authority when the national-level Gini is high (Solt, 2012). However, the objective level of income inequality has no significant impact on regime support when perceived inequality is controlled for. Compared to the significant results of subjective inequality, this result indicates that inequality perception better captures the attitudinal effects of economic inequality; the effect of economic inequality on regime support starts from people’s perception.

Results on Control Variables

Some other findings from Tables 2 and 3 are worth mentioning. First, people with a college degree or higher household income tend to have weaker demands for redistribution and weaker authoritarian value orientation; their support for the current regime is also lower. In contrast, people with a higher social status tend to have stronger authoritarian value orientation and a higher level of regime support. The contrast of these results indicates that income (or education) and social status crosscut the social cleavages in China: the high-status groups are largely distinct from the high-income (or educated) groups in political value orientation and attitudes.

Second, people with urban hukou (household registration) have stronger demands for redistribution yet weaker authoritarian value orientation; they also have lower regime support than people without urban hukou. Corresponding to these results, peasants (who mostly hold the non-urban hukou) are found reporting weaker demands for redistribution and stronger authoritarian value orientation as well as more support for the current regime. These results suggest a salient urban–rural cleavage among Chinese in terms of sociopolitical values and attitudes.

Third, it is interesting to note that CCP members have significantly weaker authoritarian value orientation than non-members do. A recent study by Ji and Jiang (2020) also finds that CCP members hold substantially more modern and progressive views than ordinary citizens on various issues including political pluralism.

Finally, perception of (upward) social mobility is positively associated with authoritarian value orientation, while perception of corruption is negatively associated with authoritarian value orientation. In fact, perceived corruption plays a similar role as perceived inequality in reducing individuals’ authoritarian value orientation and regime support. However, we find that individuals who manage to move up the social ladder are more embracive of authoritarianism and supportive of the current regime in China. Contrary to the positive effect of upward mobility on democratic support found in Latin America and sub-Saharan Africa (Houle and Miller, 2019), the above result demonstrates that high-status groups in China are more authoritarian-oriented and supportive of the regime.

Causal Mediation Effects

Results of the causal mediation analysis in Table 4 show that when perceived inequality is used as the treatment variable, the ATE of economic inequality on regime support is negative and significant, which further confirms Hypothesis 1. Apart from the indirect effect, the ADE of economic inequality on regime support is statistically significant when perceived inequality is used as the treatment variable. One such unspecified mechanisms between perceived inequality and regime support may be social trust, which is found exerting a significant effect on political attitudes in China (Huang and Schuler, 2018). When objective income inequality is used as the treatment variable, none of ATE or ADE of economic inequality is statistically significant.

When economic inequality is measured by perception, the AME of authoritarian value orientation is substantial and significant. Specifically, about 38% of the total effect of perceived inequality on regime support stems from the indirect effect operating through authoritarian value orientation. Namely, over one-third of the negative effect of perceived inequality on regime support is attributed to lower authoritarian value orientation under inequality. In contrast, the AME of redistributive preference is neither significant nor substantial. Hypothesis 2 is rejected again. When economic inequality is measured by objective income inequality, authoritarian value orientation still has a significant mediating effect, while redistributive preference does not.

Put together, causal mediation analysis results demonstrate that perceived inequality has a significant and negative effect on regime support, operating primarily through the value orientation mechanism. The redistributive preference mechanism is not empirically supported, regardless of using the subjective or objective measure of economic inequality. Moreover, objective income inequality does not affect regime support when perceived inequality is held constant. This indicates that only when inequality is perceived by individuals can it exert substantial effects on their attitudes toward the regime.

Sensitivity Analysis

Since the identification in the causal mediation analysis relies upon an untestable assumption (i.e. sequential ignorability), we need to evaluate the above results’ robustness to potential violation of the assumption. We perform sensitivity analysis that quantifies the exact degree to which the key identification assumption must be violated for the original conclusion from the causal mediation analysis to be reversed (Imai et al., 2010). The sensitivity analysis results (Figures A3 and A4 in Online Appendix) show that when perceived inequality is used as the treatment variable, for the estimated AME of authoritarian value orientation to be 0, there must be an unobserved pretreatment confounder that affects both authoritarian value orientation and regime support and causes the correlation between the error terms of the mediator and the outcome models to be greater than 0.10. This suggests that we should interpret the causal mediation mechanism with caution, as the analysis relies on a strong assumption for identification. In addition, the sensitivity analysis shows that the causal mediation effect of authoritarian value orientation is more reliable and stable than that of redistributive preference.

Robustness Tests

We conducted two sets of robustness tests for the empirical results. By and large, the empirical results, especially ones using the subjective measure of inequality, are robust in all these tests. First, we used the original 5-point scale of perceived inequality to rerun the causal mediation analysis. The results (Table A6 in Online Appendix), which are based on the specific comparison between the groups with the lowest and the highest levels of perceived inequality, are consistent with the findings using the binary variable of perceived inequality.

Second, we tested different measures of the mediators. For the mediator redistributive preference, we constructed an index of welfare-state preferences based on questions regarding respondents’ attitudes toward government’s responsibility to provide jobs, housing, health care, primary education, and pensions. We find that (Table A7 in Online Appendix) the preferences for welfare state have a significant but trivial mediating effect; moreover, this effect is insignificant when the objective measure of inequality is used.

For the mediator authoritarianism, we replaced the inclusive measure with two exclusive measures according to factor analysis and Mokken scale analysis, respectively. Factor analysis results (Tables A8 and A9 in Online Appendix) suggest that the 10 items load onto three factors, but only the first factor appears to be meaningful.¹³ Among the 10 items, items 2 and 5 have the lowest relevance or loadings (smaller than 0.2) for the first factor. Hence, we constructed an alternative index of authoritarian value orientation excluding items 2 and 5. Alternatively, the results of Mokken scale analysis (Table A10 in Online Appendix) suggest a scale that consists of all items except 2, 5, 7, and 9, from which we constructed another alternative index of authoritarian value orientation. The new results (Table A7 in Online Appendix) demonstrate that all the findings regarding the mediation effect of authoritarian value orientation still hold, and its mediating effect becomes larger.

Third, we tried an alternative dependent variable: preference for regime change.¹⁴ Intuitively, if individuals support the current regime, they will have a weak preference for regime change. Hence, we expect all the effects computed using the regime change dependent variable to be opposite to the ones using regime support as the dependent variable. This is confirmed by the new results (Table A11 in Online Appendix). Both objective and subjective inequalities are found to exert significant and positive effects on the preference for regime change, although the results with objective measure of inequality are less significant.

The second set of robustness tests adopted different model specifications. First, we conducted the causal mediation analysis using a model that includes both mediators simultaneously to compute the mediation effect of one of the mediators when the other mediator’s effect is controlled for.¹⁵ The new results (Table A12 in Online Appendix) show that our findings are robust to the inclusion of the two mediators at the same time.

Second, we employed ordered probit regressions to the outcome model and redid the causal mediation analysis. The new results (Table A13 in Online Appendix)¹⁶ show that perceived inequality increases the probability that an individual holds a lower level of regime support (support = 1, 2, or 3) while it significantly decreases their probability of holding the highest level of regime support (support = 4); moreover, only authoritarian value orientation has a significant mediating effect.

Third, we replaced the single-level regressions with multilevel ones with random intercepts varying by province and random coefficients that allow inequality’s effect to vary by province. To be noted, the multilevel regressions include provincial-level covariates such as gross domestic product (GDP) per capita, annual GDP growth rate, and government spending per capita in both mediator and outcome models. The new results (Table A14 in Online Appendix) show that the main results are robust to the multilevel modeling and inclusion of provincial-level control variables when inequality is measured by perception.

Fourth, we split the sample by survey wave and redid the causal mediation analysis for each subsample. The new results (Tables A15 and A16 in Online Appendix) show that our findings, based on the full sample, are validated in the subsamples of Wave 1 and Wave 4 as well as the aggregated sample of Waves 1, 2 and 4.¹⁷

Conclusion

Economic inequality and political legitimacy have long been thought to be connected (Lipset, 1960). It is commonly believed that China’s remarkable economic growth in recent decades has greatly contributed to its citizens’ strong support for the current regime (Chen, 2017; Ou-Yang and Zhou 2019). By contrast, less is known about the political effects of economic distribution that has been found increasingly inequitable in China. This article demonstrates that economic inequality, when perceived by individuals, has dampened their support for the current regime in China during the 2000s. In other words, the “social volcano” does exist in China (Lü, 2014a; Zhou and Jin, 2018). This finding provides a more sobering view of the economic base of political support than extant studies do. That is, economic growth and economic inequality play significant, yet distinct, roles in shaping public support for the Chinese regime. Although perceived economic growth has greatly enhanced public support for the Chinese regime, as previous research identified (Lu and Dickson, 2020; Ou-Yang and Zhou 2019), its positive effect is dampened by the negative effect that perceived inequality exerts on regime support. This finding sheds light on the motivation of the recent Chinese leaderships (e.g. Hu Jintao, Xi Jinping) to increasingly pay attention and make efforts to mitigate inequality, such as launching campaigns of poverty alleviation and social welfare expansion.

We also investigated the mechanism driving the relationship of perceived economic inequality and regime support in China. It is found that over one-third of the total effect of perceived economic inequality on regime support runs through individuals’ authoritarian value orientation. This implies that, despite continuous economic growth, if the economic distribution in China does not improve, the prospect of political instability may be looming as the masses become increasingly detached from the authoritarian values underpinning the current regime. Furthermore, we do not find evidence from China supporting the role of redistributive demand as highlighted in the political economy literature on democratization (Acemoglu and Robinson, 2006; Boix, 2003). This suggests that political mobilization (e.g. anti-government activities) in China is unlikely to be effectively built on individuals’ demands for redistribution; moreover, social divisions under economic inequality may not fall along class lines and, in all probability, cut across them.

These findings from China have generalizable implications for other East Asian countries in which economic growth often dominates public policies coupled with much less emphasis on equal or fair distributive outcomes (Holliday, 2000). The “Asian Tigers” such as South Korea and Taiwan have experienced unprecedented long-term economic growth in the post-war era, which creates a trickle-down effect on economic distribution, and the link between economic inequality and political support was considered to be “much less pronounced or even absent” (Chi and Kwon, 2016, 320–321). Drawing on evidence from China, our study partially supports, as well as disapproves, this view. We find that, on one hand, the positive effect of economic growth is greater than the negative effect of economic inequality on regime support; on the other hand, economic inequality indeed significantly changes the mindset of people and ultimately dampens regime support.

The stability and resilience of authoritarian rule have been the central focus in the “institutional turn in comparative authoritarianism” (Pepinsky, 2014). Yet to explain regime survival and stability requires us to examine how economic outcomes shape political support, especially given the widely supposed central role that economic inequality plays in democratization and regime change. This article concludes that economic distribution matters for the legitimacy and stability of an authoritarian regime, not because it provides bread (material benefits) to the people, but because it sustains people’s identification with the core political values that underpin the regime. In this way, our study returns attention, regarding authoritarian legitimacy and stability, to political values that play an important role in shaping political support. Extant studies of authoritarian regimes focus mostly on the ruling strategies and institutional characteristics. The role of political value orientation in shaping popular support for authoritarian regimes has been overlooked and deserves more future research.

Supplemental Material

sj-docx-1-psx-10.1177_00323217211048040 – Supplemental material for Bread or Roses: How Economic Inequality Affects Regime Support in China?

Supplemental material, sj-docx-1-psx-10.1177_00323217211048040 for Bread or Roses: How Economic Inequality Affects Regime Support in China? by Xian Huang and Cai Zuo in Political Studies

Supplemental Material

sj-docx-2-psx-10.1177_00323217211048040 – Supplemental material for Bread or Roses: How Economic Inequality Affects Regime Support in China?

Supplemental material, sj-docx-2-psx-10.1177_00323217211048040 for Bread or Roses: How Economic Inequality Affects Regime Support in China? by Xian Huang and Cai Zuo in Political Studies

Footnotes

Acknowledgements

The authors are grateful to the three anonymous reviewers who provided valuable suggestions for the revision of this article. We thank Dan Chen, Ling Chen, Iza Ding, Jean Hong, Xiaobo Lü, Haruka Nagao, Min Tang, Jessica Teets, Zhengxu Wang for their useful comments. An earlier version of this article was presented at the 2020 American Political Science Association, 2021 Midwest Political Science Association, and 2021 Southern Political Science Association annual conferences. All errors remain our own.

Declaration of Conflicting Interests

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

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Xian Huang

Cai Zuo

Supplementary Information

Additional Supplementary Information may be found with the online version of this article.

Online Appendix: Supplementary Figures.

Figure A1. Causal Mediation Analysis Results by Sample (Mediator: Redistributive Preference).

Figure A2. Causal Mediation Analysis Results by Sample (Mediator: Authoritarian Values).

Figure A3. Sensitivity Analysis for the Causal Mediation Effect with Redistributive Preference as Mediator.

Figure A4. Sensitivity Analysis for the Causal Mediation Effect with Authoritarian Value Orientation as Mediator.

Online Appendix: Supplementary Tables.

Table A1. Provincial Distribution of Observations.

Table A2. Results of Factor Analysis on Regime Support Items.

Table A3. Results of Factor Analysis on Regime Support Items.

Table A4. Results of Mokken Scale Analysis on Regime Support Items.

Table A5. Summary Statistics of Variables.

Table A6. Results of Causal Mediation Analysis Using Five-Point Scale of Perceived Inequality.

Table A7. Results of Causal Mediation Analysis with Alternatively Measured Mediators.

Table A8. Results of Factor Analysis on Authoritarian Value Items.

Table A9. Results of Factor Analysis on Authoritarian Value Items.

Table A10. Results of Mokken Scale Analysis on Authoritarian Value Items.

Table A11. Results of Causal Mediation Analysis with Preference for Regime Change as the Dependent Variable.

Table A12. Results of Causal Mediation Analysis with Multiple Mediators.

Table A13. Results of Causal Mediation Analysis Using Ordered Probit Regression for the Outcome Model.

Table A14. Results of Causal Mediation Analysis Using Multilevel Regressions with Provincial Control Variables.

Table A15. Results of Causal Mediation Analysis by Year (Independent Variable: Perceived Inequality).

Table A16. Results of Causal Mediation Analysis by Year (Independent Variable: Income Inequality).

Notes

Author Biographies

Xian Huang is an Assistant Professor of Political Science at Rutgers University. Her research has focused on the politics of social inequality and redistribution with a regional focus on China and East Asia. Her research has appeared in Governance, Studies in Contemporary International Development, and Social Science Research, among other. Her recent book is Social Protection under Authoritarianism: Health Politics and Policy in China (Oxford University Press, 2020).

Cai Zuo is an associate professor at School of International Relations and Public Affairs, Fudan University. Her research focuses on comparative political institution, political methodology, bureaucratic politics, and Chinese politics. Cai earned her doctoral degree in political science from the University of Wisconsin–Madison. Her research appeared in International Political Science Review, European Political Science, China Quarterly, and Journal of Contemporary China.

References

Acemoglu

Robinson

(2006) Economic Origins of Dictatorship and Democracy. New York: Cambridge University Press.

Alesina

La Ferrara

(2005) Preferences for Redistribution in the Land of Opportunities. Journal of Public Economics 89 (5–6): 897–931.

Alesina

Cozzi

Mantovan

(2012) The Evolution of Ideology, Fairness and Redistribution. The Economic Journal 122 (565): 1244–1261.

Anderson

(2007) The End of Economic Voting? Contingency Dilemmas and the Limits of Democratic Accountability. Annual Review of Political Science 10: 271–296.

Anderson

Singer

(2008) The Sensitive Left and the Impervious Right: Multilevel Models and the Politics of Inequality, Ideology, and Legitimacy in Europe. Comparative Political Studies 41 (4–5): 564–599.

Ansell

Samuels

(2014) Inequality and Democratization. New York: Cambridge University Press.

Ariely

Uslaner

(2017) Corruption, Fairness, and Inequality. International Political Science Review 38 (3): 349–362.

Bavetta

Li Donni

Marino

(2020) How Consistent Are Perceptions of Inequality. Journal of Economic Psychology 78: 102267.

Beloshitzkaya

(2020) Democracy and Redistribution: The Role of Regime Revisited. East European Politics and Societies and Cultures 34 (3): 571–590.

10.

Benabou

(2001) Social Mobility and the Demand for Redistribution: The POUM Hypothesis. Quarterly Journal of Economics 116 (2): 447–487.

11.

Boix

(2003) Democracy and Redistribution. Cambridge: Cambridge University Press.

12.

Bratton

Mattes

(2001) Support for Democracy in Africa: Intrinsic or Instrumental? British Journal of Political Science 31 (3): 447–474.

13.

Chen

(2017) Local Distrust and Regime Support: Sources and Effects of Political Trust in China. Political Research Quarterly 70 (2): 314–326.

14.

Chi

Kwon

(2016) The Trust-Eroding Effect of Perceived Inequality: Evidence from East Asian New Democracies. The Social Science Journal 53: 318–328.

15.

Chu

Y-H

(2013) Sources of Regime Legitimacy and the Debate Over the Chinese Model. The China Review 13 (1): 1–42.

16.

Chu

Y-H

C-E

(2020) Equality of Opportunity and Satisfaction with the Political System: The Mediating Role of Regime Type. In: Park

C-M

Uslaner

(eds) Inequality and Democratic Politics in East Asia. New York: Routledge, pp.96–115.

17.

Diamond

Linz

Lipset

(1988) Democracy in Developing Countries, vol. 1. Boulder, CO: L.Rienner.

18.

Finseraas

(2009) Income Inequality and Demand for Redistribution: A Multilevel Analysis of European Public Opinion. Scandinavian Political Studies 32 (1): 94–119.

19.

Gilley

(2006) The Determinants of State Legitimacy: Results for 72 Countries. International Political Science Review 27 (1): 47–71.

20.

Gimpelson

Treisman

(2018) Misperceiving Inequality. Economics & Politics 30 (1): 27–54.

21.

Haggard

Kaufman

(2012) Inequality and Regime Change: Democratic Transitions and the Stability of Democratic Rule. American Political Science Review 106 (3): 495–516.

22.

Haggard

Kaufman

Long

(2013) Income, Occupation, and Preferences for Redistribution in the Developing World. Studies in Comparative and International Development 48 (2): 113–140.

23.

Hetherington

(2005) Why Trust Matters. Princeton, NJ: Princeton University Press.

24.

Holliday

(2000) Productivist Welfare Capitalism: Social Policy in East Asia. Political Studies 48 (4): 706–723.

25.

Houle

(2018) Does Economic Inequality Breed Political Inequality? Democratization 25 (8): 1500–1518.

26.

Houle

Miller

(2019) Social Mobility and Democratic Attitudes: Evidence from Latin America and sub-Saharan Africa. Comparative Political Studies 52 (11): 1610–1647.

27.

Huang

K-P

Schuler

(2018) A Status Quo Theory of Generalized Trust: Why Trust May Reduce the Prospects for Democratic Transition in East Asia. Comparative Politics 51 (1): 121–139.

28.

Huang

Gao

(2017) Does Social Insurance Enrollment Improve Citizen Assessment of Local Government Performance? Evidence from China. Social Science Research 70: 28–40.

29.

Huhe

Tang

(2017) Contingent Instrumental and Intrinsic Support: Exploring Regime Support in Asia. Political Studies 65 (1): 161–178.

30.

Imai

Keele

Yamamoto

(2010) Identification, Inference and Sensitivity Analysis for Causal Mediation Effects. Statistical Science 25 (1): 51–71.

31.

Imai

Keele

Tingley

, et al. (2011) Unpacking the Black Box of Causality: Learning about Causal Mechanisms from Experimental and Observational Studies. American Political Science Review 105 (4): 765–789.

32.

Jiang

(2020) Enlightened One-Party Rule? Ideological Differences between Chinese Communist Party Members and the Mass Public. Political Research Quarterly 73 (3): 651–666.

33.

Jost

Hunyady

(2002) The Psychology of System Justification and the Palliative Function of Ideology. European Review of Social Psychology 13 (1): 111–153.

34.

Knutsen

Wegmann

(2016) Is Democracy about Redistribution? Democratization 23 (1): 164–192.

35.

Krieckhaus

Son

Bellinger

, et al. (2013) Economic Inequality and Democratic Support. The Journal of Politics 76 (1): 139–151.

36.

Kuhn

(2013) Inequality Perceptions, Distributional Norms, and Redistributive Preferences in East and West Germany. German Economic Review 14 (4): 483–499.

37.

Lai

D-Y

Lue

J-D

W-C

(2021) Intergenerational Mobility and Preference for Redistribution: Evidence from East Asia. Journal of Asian Public Policy 14 (1): 45–62.

38.

Lewis-Beck

Stegmaier

(2000) Economic Determinants of Electoral Outcomes. Annual Review of Political Science 3: 183–219.

39.

Gong

Xiao

(2016) The Perception of Anti-corruption Efficacy in China: An Empirical Analysis. Social Indicators Research 125: 885–903.

40.

Lipset

(1960) Political Man: The Social Bases of Politics. Garden City, NY: Anchor Books.

41.

Dickson

(2020) Revisiting the Eastonian Framework on Political Support: Assessing Different Measures of Regime Support in Mainland China. Comparative Politics 52 (4): 671–701.

42.

Lü

(2014a) Does Changing Economic Well-Being Shape Resentment about Inequality in China? Studies in Comparative International Development 49 (3): 300–320.

43.

Lü

(2014b) Social Policy and Regime Legitimacy: The Effects of Education Reform in China. American Political Science Review 108 (2): 423–437.

44.

Lü

Scheve

(2016) Self-Centered Inequity Aversion and the Mass Politics of Taxation. Comparative Political Studies 49 (14): 1965–1997.

45.

Meltzer

Richard

(1981) A Rational Theory of the Size of Government. Journal of Political Economy 89 (Oct): 914–927.

46.

Osberg

Smeeding

(2006) Fair Inequality? Attitudes toward Pay Differentials: The United States in Comparative Perspective. American Sociological Review 71: 450–473.

47.

Ou-Yang

Zhou

(2019) Economic Evaluations and Political Support in Authoritarian Countries: A Comparative Viewpoint. Asian Survey 59 (4): 738–768.

48.

Park

C-M

(2020) Assessing the Inequality-Democracy Linkage at the Individual Level: Evidence from East Asia. In: Park

C-M

Uslaner

(eds) Inequality and Democratic Politics in East Asia. London; New York: Routledge, pp.75–95.

49.

Pepinsky

(2014) The Institutional Turn in Comparative Authoritarianism. British Journal of Political Science 44 (3): 631–653.

50.

Perrella

AML

Belanger

Nadeau

, et al. (2016) Does a Growing Income Gap Affect Political Attitudes? Canadian Public Policy 42 (1): 35–48.

51.

Pop- Eleches

Tucker

(2017) Communism’s Shadow: Historical Legacies and Contemporary Political Attitudes Princeton Studies in Political Behavior. Princeton, NJ: Princeton University Press.

52.

Riker

(1990) Political Science and Rational Choice. In: Alt

Shepsle

(eds) Perspectives on Positive Political Economy. Cambridge: Cambridge University Press, pp.163–181.

53.

Rose

Mishler

(2002) Comparing Regime Support in Non-democratic and Democratic Countries. Democratization 9 (2): 1–20.

54.

Rose

Mishler

Munro

(2011) Popular Support for an Undemocratic Regime: The Changing Views of Russians. New York: Cambridge University Press.

55.

Shi

(2015) The Cultural Logic of Politics in Mainland China and Taiwan. New York: Oxford University Press.

56.

Solt

(2008) Economic Inequality and Democratic Political Engagement. American Journal of Political Science 52 (1): 48–60.

57.

Solt

(2012) The Social Origins of Authoritarianism. Political Research Quarterly 65 (4): 703–713.

58.

Tang

Huhe

Zhou

(2015) Contingent Democratization: When Do Economic Crises Matter? British Journal of Political Science 47: 71–90.

59.

Tian

(2012) Sheng Yu Jumin Shouru Jini Xishu Cesuan Ji Qi Biandong Qushi Fenxi [Provincial Income Gini Coefficient Estimate and Changing Trend Analysis]. Jingji Kexue [Economics] 2: 48–59.

60.

Tingley

Yamamoto

Hirose

, et al. (2014) Mediation: R Package for Causal Mediation Analysis. Journal of Statistical Software 59 (5): 38.

61.

Visser

Lubbers

Kraaykamp

, et al. (2014) Support for Radical Left Ideologies in Europe. European Journal of Political Research 53: 541–558.

62.

Wang

(2005) Before the Emergence of Critical Citizens: Economic Development and Political Trust in China. International Review of Sociology 15 (1): 155–171.

63.

Welzel

Inglehart

(2009) Political Culture, Mass Beliefs, and Value Change. In: Haerpfer

Bernhagen

Inglehart

, et al. (eds) Democratization. New York: Oxford University Press, pp.126–144.

64.

Whyte

(2010) Myth of the Social Volcano: Perceptions of Inequality and Distributive Injustice in Contemporary China. Stanford, CA: Stanford University Press.

65.

(2009) Income Inequality and Distributive Justice: A Comparative Analysis of Mainland China and Hong Kong. The China Quarterly 200: 1033–1052.

66.

Xie

Zhou

(2014) Income Inequality in Today’s China. Proceedings of the National Academy of Sciences of the United States of America 111 (19): 6928–6933.

67.

Zhao

(2020) Visible and Invisible Hands Intertwined: State-Market Symbiotic Interactions and Changing Income Inequality in Urban China. Social Science Research 91: 102450.

68.

Zhou

Jin

(2018) Inequality and Political Trust in China: The Social Volcano Thesis Re-examined. The China Quarterly 236: 1033–1062.

69.

Zhu

Huang

Zhang

(2019) “Big Tigers, Big Data”: Learning Social Reactions to China’s Anticorruption Campaign through Online Feedback. Public Administration Review 79 (4): 500–513.

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.

0.00 MB

0.43 MB

0.28 MB