The Paradox of Middle-Class Attitudes in China: Democracy,Social Stability,and Reform

Abstract

This article explores the seemingly paradoxical attitudes of the Chinese middle class towards democracy, social stability, and reform. Using fieldwork data from Ningbo, this article shows that a group of objective, middle-class individuals can concurrently display high levels of support for democratic principles and low levels of participation in real-life socio-political events. Being generally confident in China's social stability, these individuals have little to no desire for significant democratic reform, or indeed any reform that occurs outside the purview of the state, as it is considered destabilising. By highlighting the distinction between how these members of the middle class respond to generic democratic concepts, real-life sociopolitical affairs, and the idea of democratic reform, this article argues that the Chinese middle class are aware of what “should be,” what “could be,” and what “is,” which lends their socio-political attitudes a paradoxical appearance.

Introduction

Despite being ascribed important modernising functions in history (Fukuyama and Bloom 1989; Easterly 2001), the political outcome of the middle class in developing nations is a contested one. One of the most debated arenas is in China, where the increasing number of affluent city-dwellers have been described both as “not afraid to get political” in scholarly and comment pieces alike, (Li 2003; Tang 2011; Fukuyama 2013; Wasserstrom 2008) and as having a keen vested interest in the state and the existing socio-political order (Chen 2013; Tomba 2009; Goodman 2014). Moreover, there is considerable debate on the size of the Chinese middle class in terms of who should be included and, indeed, whether one exists at all (Goodman 2012; Guo 2008). Due to differences in definitions and observational angles, the Chinese middle class appear to exhibit varying, sometimes paradoxical political attitudes (Tang, Woods, and Zhao 2009; Sonoda 2010; Li Chunling 2010; Li Cheng 2010).

One of the key reasons behind such discrepancies is that these studies are, by and large, top-down and macro-oriented. In order to draw these conclusions, the middle class need to be viewed as a largely homogeneous group, sharing similar collective interests arising out of their similar socio-economic positions. The factors which influence their attitudes and behaviour are largely external: their socio-economic circumstances (how they became “middle”) and their socio-political freedoms (whether they can act upon their “middle-ness”). Yet, this is a gross simplification of matters. Even without digressing into the politically charged and intellectually varied debate on the definition and relevance of class, it is not difficult to imagine that the answer to the question What do the Chinese middle class think? is not a simple binary one. In fact, the key question here is not so much what they think, but why, and more importantly, how their attitudes might vary in different contexts. Previous studies on middle-class attitudes have, by and large, focused their questions and surveys on abstract concepts such as liberty, rights, and democracy, or general questions on the nature and extent of middle-class political participation without qualifying, or delving into, the details of what people think they will do, or what they have done, in the context of specific events, or how people react to certain topics. This makes sampled answers prone to social desirability bias, and often results in mismatches between abstract and concrete thinking.

This article seeks to overcome the current state of binary analysis in middle-class political studies by offering a fresh look at the differences in middle-class attitudes towards abstract concepts of democracy, real-life socio-political events that require mass participation, and the prospect of democratic reform. Using the city of Ningbo as a case study, this article examines a group of urban, affluent individuals who fit the conventional middle-class criteria and would otherwise be considered part of the broad category of the “middle class.” A total of 182 individuals were questioned via Internet survey about their attitudes towards abstract political principles and then about their participation in the Zhenhai PX incident, the largest mobilisation of ordinary citizens around a politically charged topic the city has witnessed in decades. A further 19 individuals were drawn from the respondent pool and interviewed about their views towards China's socio-political stability and the prospect of democratic reform as a whole. In response, these participants exhibited a high level of support for democratic principles, but a low level of desire for radical political reform or indeed any reform outside the purview of the state. Rather than reverting to, and depending fully on, the “vested interest” thesis, this paper highlights the distinction between respondents’ attitudes towards generic political concepts, real-life socio-political events, and the actual prospect of political reform and shows that these members of the urban affluent are particularly adept at differentiating between what “should be,” what “could be,” and what “is.” Due to their education and socio-economic position, they are typically aware of and accepting of democratic concepts; however, they also (believe themselves to) have a sound knowledge of the socio-political environment at large and are aware of the distinction between the desirability of a concept and the feasibility of its implementation. This distinction in their attitudes can serve to shed light on the seemingly contradictory attitudes of the Chinese middle class presented in previous literature: they are both able to recognise the merits of democratic reform in principle and express caution in its real-life endorsement. This cautious attitude represents neither a simple deference to, nor a fear of, the Chinese state, but rather a calculated response to the current status quo (stability) versus the risks of reform (instability). As they remain observant of China's socio-political setting, their relationship with the Chinese state is likely to be integrative and reciprocal, rather than simply oppositional or subservient.

Research Methods

In contrast to most middle-class studies, this article does not focus on inter-class comparisons across a representative sample of the population, but instead examines a section of the middle-class population to explore their political attitudes from several angles. Adjustments were thus made in the data-collection method: for questions on general political principles and the PX incident, 182 respondents were recruited via an Internet survey, not only because anonymity encourages people to participate, but also because the requirement of Internet connectivity automatically excludes those in unconnected rural regions, who are unlikely to be middle class. The respondent locale was limited to Ningbo via both IP monitoring and Ningbo-specific questions. Snowball-sampling methods were used to maximise response: respondents were more likely to receive the survey through their expanded social network (deemed more trustworthy), and were allowed to retain their anonymity should they choose to participate. In the initial stages, the survey was distributed to the online messenger groups of a large private corporate enterprise, a medium-sized real-estate firm, a medium-sized insurance firm, and a regional car owners club.

Since the purpose of this article is not to argue the finer details of what construes middle class, respondents were considered middle class if they fulfilled one or more of three conventionally defined, objective criteria: household average income (2.5 times above the city average: approximately CNY 87,500 per year [Li and Zhang 2008]), occupation (white-collar occupations, as outlined in Lu Xueyi's 10 occupational strata [Lu 2002; Chen 2013]), and education (college-level and above). The point is to show that the sampled population can be, and often are, referred to as “China's rising middle class,” but they do not necessarily exhibit either of the binary attitudes presented in previous studies. There are a multitude of cognitive processes at work that influence respondents’ political attitudes, some of which are unrelated to their class positions or even class identity. Indeed, subjective class identity, while also gathered, did not appear to make a significant difference to respondents’ political attitudes. This further illustrates the need to consider the context of respondents’ political attitudes, rather than simply viewing the sample population as necessarily sharing any coherent identity or attitudes based on their objective status. The final valid sample consisted of 182 objectively middleclass respondents (32 per cent were state employees, 46 per cent were men, 31 per cent were CCP members, and 46 per cent subjectively identified themselves as middle class). Of these, 19 respondents who satisfied all three objective criteria (nine were male, ten identified themselves as middle class, six were 20–29 years old, five were 30–39 years old, four were 40–49 years old, and four were 50+ years old) agreed to be drafted anonymously for semi-structured interviews using questions about China's political stability and reform.

Attitude towards Democratic Concepts

In order to uncover layered reasoning behind middle-class attitudes and behaviour, this study divides the lines of questioning on three levels. First, respondents were asked for their opinions on abstract political concepts similar to those found in large national survey samples. These statement questions derived from the East Asian Barometer and the results are presented in Table 1.

Table 1

Middle-Class Attitudes towards Democratic Concepts

Statement	Positive response percentage 100% = 182
Harmony of the community will be disrupted if people organise lots of groups.	26.7
Measures to promote political reform should be initiated by the Party and government, not by ordinary people (, lao baixing) like me.	60.0
Government leaders are like the head of a family; we should all follow their decisions.	18.3
Political elections should have multiple, diverse candidates.	88.9
Competition among several parties in election of government leaders should not be allowed.	20.6

The general trend here is clear: when assessed as an objective class category, the respondents of this sample appear to be considerably liberal in their political attitudes. Their overwhelming support for the concept of civil society and multiparty elections is particularly notable. While support for multi-candidate elections is commonplace, as it is already in practice in China, the idea of multiparty elections is usually met with caution, as China is a definitively one-party state, and to support anything else could be interpreted as a challenge to the current regime. Before we draw the conclusion that these members of the Chinese middle class appear to have matured in their democratic outlook, however, we should pay attention to the idiosyncrasies presented in this sample: for example, 60 per cent of respondents would still prefer reform to be led by the government, in spite of their otherwise “democratic” outlook. This deviation is most significant because it presents itself as a major anomaly; it appears to be incompatible with respondents’ lack of support for the corresponding statement that “government leaders are like the head of a family,” which nearly 82 per cent of respondents in this sample did not agree with. In their responses to these two statements, those sampled seem to have two opposing preferences: it appears that they are simultaneously in favour of and against authoritarian leadership. What could have caused such an anomaly? In order to put these results into context, respondents were then questioned about the Zhenhai PX protest, as a case study of their real-life socio-political participation.

Case Study: Zhenhai PX Protest

From 24 October through 28 October 2012, a series of protests erupted in Ningbo over the proposed expansion of a petrochemical plant to produce the industrial chemical paraxylene (PX) in the district of Zhenhai, a mere 15.5 kilometres away from the city centre. Paraxylene is believed to have severe adverse effects on health, and associated projects have been highly controversial nationwide. Other PX-related projects have been rejected in the cities of Dalian and Xiamen, due to similar protests over health and environmental risks. Emotional pleas and propaganda filled the Internet for weeks leading up to the protest, peaking on the weekend of 27/28 October. The following day, on 29 October, the Ningbo Municipal Government conceded and announced that the project would be halted indefinitely, pending further review.

Although the media (especially the foreign media) were eager to portray the incident as a “middle-class uprising” (Larson 2012; French 2006), China Daily stated that the protests were not inherently a rebellion against the government or even the project, but that the main grievance arose from a lack of information about the project and government accountability for it (Yang 2010). Nevertheless, engaging in such an event of mass mobilisation, as well as being in direct conflict with the authorities, is inherently political, thus this article's enquiry into the nature of and participation in the PX incident is very telling of how respondents approached real-life socio-political events.

A series of multiple-choice questions were given to the respondents, who were asked to pick one or more statements that best matched their attitude towards and participation in the PX incident. Overall awareness of the incident was quite high: 65 per cent of all surveyed participants were interested in it. However, actual participation was very low: only 25 per cent participated online and a mere 6 per cent turned up at a protest or participated in some way offline. This is consistent with previous survey findings that suggest that the middle class have high civic awareness, but low civic association (Wang 2008). No significant correlation was found between respondents’ age, subjective class identity, and their participation. CCP membership and state employment status had a predictable effect on respondents’ participation: those who had closer ties with the state apparatus were less likely to participate in such politically sensitive events. When examined against their answers to the previous question set on abstract political concepts, it appears that respondents’ proclaimed political attitudes did have a degree of influence on their political behaviour, though not tremendously: a larger proportion of those who believed civil society to be disruptive felt disinterested in the PX incident as a whole, while those who saw the government as being like the “head of a family” were generally more disinclined to get involved, whether through discussion or online participation, by a margin of 15 per cent (see Appendix). More interestingly, those who believed that the government should lead reforms were more likely to participate online, but steered away from offline protest, while those who did not believe the government had to lead reforms engaged in more offline activities. This suggests that those who believed in state-led reform initiatives saw online participation both as a legitimate outlet for their concerns and as being within the purview of the state, as the Internet is censored and monitored. Offline participation, by contrast, was seen to be an extra-state activity, and was thus less condoned by those who believe that the state has an essential role to play in the political process. Of course, the numbers of people who turned up at the protests were nevertheless low. Due to the overwhelming support for multi-candidate elections, the high standard deviation presented for the statement “I participated online in the PX incident” is more of an outlier on account of the small sample size, rather than any strong correlation.

The results of the semi-structured interviews with the smaller sample of respondents are more revealing, particularly in terms of how the interviewees remembered and explained the PX incident using their own rationale. None of the interviewed respondents believed the incident was an exclusively middle-class affair or that the middle class had played a prominent role in the events. Typically, they argued that people participated in the demonstrations because “a line had been crossed”: PX, believed to be highly toxic, was seen as “threatening people's lives, regardless of class.” Often, they described the struggle as being between those who had a vested interest in the expansion of the chemical plant and those who had a vested interest in the safety and health of the local area. Some even thought that the middle class would be the least relevant party, since they would be able to move out of the area, leaving behind the poor as the most vulnerable to a polluted environment. One respondent argued that, if it were not for the fact that public health was at stake, people would never have taken to the streets to demonstrate against the plant.

The imbalance between the potential repercussions from and rewards for participation was a major factor in interviewees’ decisions to opt out of the protests. CCP members and the employees of state-owned enterprises and government institutions recalled how warnings against participating in the demonstrations were distributed to them through mass texts and departmental bulletins. Others stressed that they had a “normal life” at home, which they did not want to put at risk. Although they demonstrated a reasonably good understanding of popular political topics, these respondents preferred to keep their activities in the private sphere. Some respondents said that although they understood that those who took to the streets felt “their lives being threatened,” they did not share this sentiment, so they themselves declined to participate.

During the interviews, it became clear that the interviewees had various different rationales to explain other peoples’ “active” behaviour and their own “passive” stance. Several respondents further identified the root of the problem as a lack of information dissemination and transparency about the PX plant, but did not feel the protest had legitimate environmental grounds. They expressed compassion and understanding of the protesters’ cause, but were able to analyse the issue beyond the superficial call to rally. Some respondents believed that the protesters had been misled, not in terms of the outcome of the protest, but in terms of the reasons for participating in the protests. One respondent argued that media exposure had put an unfortunate spotlight on PX, while many other more polluting chemical projects were being left unexamined. Though the sentiments and motivation behind the PX demonstrations were generally endorsed by respondents, they made a distinction between “understanding” and “participating.” Many indicated that they personally would not have followed the protesters to the demonstrations, as they felt the risks of participation outweighed the benefits.

It was clear that these respondents could readily acknowledge and accept the popular rationale behind the protests, but they also prided themselves in “being able to see beyond it.” As PX is believed to have adverse health effects, they could see that the public's reaction was to be expected; however, because they believed that the adverse health effects of PX had been greatly exaggerated, many respondents felt the cause for protest had been sensationalised and they, therefore, saw the project as less of a threat than did the active protest participants. Instead, because they saw the struggle between industrial and commercial interests over this project, they wanted to disassociate themselves from anyone who appeared “rash” or “hotheaded,” as they felt that these individuals were prone to manipulation. Even though these middle-class respondents understood that protesters felt that “a line had been crossed,” for them, that line was still at some safe distance. Therefore, the costs of participating in this highly politically charged protest would outweigh the benefits. Nevertheless, the fact that they could sympathise with the protesters suggests that if, one day, that invisible line were to be crossed, then they too might join others in open protest.

Middle Class, Social Stability, and Reform

Finally, in order to give more context to the results gathered above, respondents were asked to assess China's current level of social stability and their own opinions on the role the middle class play in maintaining or affecting such stability. Due to their sensitive nature, these questions could not be included in the larger questionnaire survey, but instead were asked exclusively in the more intimate setting of interviews.

When asked to evaluate China's present level of social stability, most respondents placed their assessment on the stable side of the scale – that is, they generally allocated a score from 6 to 8, where 1 denotes complete instability and 10 represents total stability. It was unanimously agreed that Ningbo is more stable than China as a whole, and respondents gave the city a score of at least 8 or 9 on the stability scale. The mild temperament of local residents and their propensity for undertaking money-making activities rather than expressing political ambitions, together with the prosperity of the region and its lack of natural disasters, were the primary reasons for respondents’ high levels of confidence in Ningbo's stability, both at the present and in the long term. Of course, it is likely that, as they also belong to the population of Ningbo, these respondents might also ascribe similar favourable characteristics to themselves and also might subscribe to the same rationale for their own reticence towards political participation and reform.

Generally speaking, respondents said they believed that the middle class would be stabilisers of society, rather than agents of change. This is partly due to their own casual observation of Western middle-class societies and partly due to their own experience. Those who could remember a time of poverty often made comparisons with their past, arguing that they were the beneficiaries of societal change and thus would not endanger the very socio-political system that had brought about those benefits. The younger respondents were similarly aware of the benefits that a state-sector job would bring. One respondent related a story where even his most “rebellious” friend, someone who had often spoken against the state, joined the civil service at the first opportunity upon graduation. Thus, many respondents expressed a certainty that the middle class would not be partial to extensive and uprooting reforms, due to their high living standards and state policies aimed at promoting these lifestyles. As found in many other academic works, they argued that the middle class would have a vested interest in maintaining the status quo, since they are the “haves” of present-day society.

None of the respondents subscribed to the idea of a “middle-class awakening,” whereby the middle class would develop political demands on top of their socio-economic stability; instead, they argued that any motivation for change would come from defensive desperation, rather than proactive demands. The phrase “If you are poor you will think about change” (, qiongze sibian) came up frequently during discussions, bringing to mind William Cobbett's famous quote, “You cannot agitate a man on a full stomach.” Hence, many respondents praised China's current efforts at reform and said that they believed these efforts would appease the minority within the middle or lower-middle classes who are not entirely happy with the status quo. If not, they joked, the middle classes can always “vote with their feet,” and emigrate to another country, where their political demands would be better met. Open calls for reform were seen as confrontational and the “worst strategy possible,” since they would most likely prompt an adverse reaction from the state. Respondents’ overall confidence in state-led reforms was reasonably high. Respondents did not typically see the state as oppressive or autocratic. They emphasised that the leadership also sought stability and would allow for necessary reforms to maintain stability. Thus, it could be said that respondents’ goals and interests were aligned with the state, which they saw as a leader, as well as an ally, rather than an opponent. Indeed, the current leadership's propensity for reform appears to boost respondents’ confidence in China's long-term stability:

I don't think there will be any major instability and changes in China. When I was younger, I thought there was a good chance of socio-political turbulence, but now I don't think so anymore. Especially with the new leadership, they are all talking about reform, and I think that's great. Once you reform, you avoid the chances of a major upheaval. I think the new leadership knows this and this is their subtle way of saying, we need to change. (Anonymous 1 2013)

By extension, some warned that if the state undermines middle-class interests or if the middle class believe that the state is undermining their interests, then this mutually beneficial partnership might fall apart. The shrinking and devaluing of the stock market and financial products, together with skyrocketing prices in the housing market, argued some, are already putting pressure on the middle classes. With their savings devaluing in the banks and real estate becoming increasingly unaffordable, the very basis of the middle class – income and property, in the eyes of many – is threatened, along with the prospects for stability. Nevertheless, the same respondents hastened to add that they themselves would not hope for instability or change, since it is always the lao baixing (, the ordinary people), the overarching term with which respondents primarily identified – rather than referring to their class – who lose out in times of flux.

Indeed, although some respondents were wary of the possibility of instability arising because the collective interests of the middle class had been undermined, none of the interview respondents believed that they themselves would be the instigators of reform. Respondents’ personal attitudes towards socio-political change were, in fact, very reserved. Respondents typically took a stance of “passive acceptance,” claiming that, while they were aware that the status quo was not perfect, there was very little that could ostensibly be changed. Their search for “better alternatives” had so far drawn a blank, and any prospect of socio-political change was riddled with risks and unpredictable outcomes, whether imagined or real. One respondent drew a particularly vivid image:

To go for reform now is like standing in a dark room with no light. Now you might be told there is a light somewhere in the room, but there are a hundred switches and only one of them is real, the others electrocute you. How would you move then? Of course you stay in the dark and stay where you are. (Anonymous 2 2012)

The younger respondents were particularly quick to disassociate themselves from “class consciousness” of any kind that might have a socio-political impact, while the middle-aged respondents highlighted their preference for focusing on their private lives, thus making them disinclined towards change. A distinction was made frequently between those people who “keep their heads down” and “go about their own lives,” a group of people with whom many of those interviewed identified, and the “angry youth” (, fen qing), the “50-cent army” (, wu mao: a slang term for Internet commentators hired by the state to post favourable comments about party policy, so-called for the 50 fen (0.5 CNY) they allegedly receive for each post), and the “American saboteurs” (, mei fen: a slang term for Chinese netizens who “worship” the United States and are overly critical of China) who are active primarily on the Internet. The former group was seen as rational, calm, and unswayed by sensational headlines. They were also seen as not necessarily apathetic, but better informed. The latter were instead criticised for being rash and eager and for “rushing to the forefront of public attention whenever injustice arises,” thus being prone to manipulation.

The older respondents saw the notions of reform and stability through the lens of their own life experiences. For them, the terms “political participation” and “reform” resonated with specific, historical meaning, which led them to believe that a recurrence of these themes was not possible, or indeed desirable, anytime soon. Even though the phrase introduced alongside the term “political participation” in the discussion was “reform” (, gaige), many elderly respondents substituted this phrase with the words for “revolution” (, geming), a term which they then used interchangeably with “reform.” This suggests that the older respondents’ idea of active political participation was a more radical and violent one, echoing the events of their childhood. This negative connotation was likely to discourage them further from supporting any future calls for reform.

Meanwhile, a small number of respondents exhibited a certain level of pragmatism vis-à-vis the idea of proactive reform. They stressed that they would not initiate any demands for reform or change, but would evaluate any opportunities for reform should they arise. They said that action was warranted only if the timing, need, and leadership were right, in accordance with the Chinese saying “opportune time, advantageous terrain, and popular support” (, tianshi dili renhe); therefore, action was not required if the situation was “just a fussy scandal of no consequence.” Far from being the initiators of reform and active pursuers of change, these middle-class respondents preferred to take the role of watchful bystanders, lest their actions destabilise the status quo.

Together with their acute awareness of the existing opportunities and pitfalls in the current economic climate and the acknowledgement of their relative success under such a climate, the respondents’ tentative attitude towards change and their hope for continued betterment are perhaps best illustrated by one respondent's reference to the famous opening line of A Tale of Two Cities, an epic story also set in a time of great transformation:

People like me, I wouldn't want instability, that'd be good for no one. Honestly, I think it's like Dickens said, “It was the best of times, it was the worst of times.” Right now, we can really make something for ourselves if we try, the opportunities are there. But society is evolving so fast, if you can't grasp the opportunities or jumped on the wrong wagon, you'd end up being miserable. I don't like change. If you look at the films and TV shows depicting great eras of change, they are all bloody and turbulent. (Anonymous 2 2013)

Discussion and Conclusion

This article presents a multilayered look into the socio-political attitudes of a group of urban, affluent respondents in Ningbo, who by conventional criteria can be called “middle class.” The results of this research reveal a divergence in respondents’ political orientation and their tendency towards actual political participation, as well as their reaction towards the prospect of democratic reform. On the one hand, when asked in general terms, the respondents showed strong support for political liberty, competitive elections, and participatory norms, all signs that might point to the “democratisation of the middle class.” On the other hand, however, when the topic of the PX incident was introduced as a specific case study, the level of proactive participation was very low and actual physical participation in the demonstrations was almost negligible. Furthermore, when asked to consider the prospect of democratic reform in China and their role as “middle class” within it, respondents almost always disassociated themselves from the reformative element and believed instead that they were the social stabilisers.

This apparent paradox can be approached from several angles. First, it is worth reiterating that like other studies on this topic, the social group sampled here, despite fitting the conventional middle-class criteria, were far from a homogeneous entity. The most prominent holding factor for respondent attitudes towards democratic concepts were, predictably, CCP membership and state employment status. This concurs with a wealth of previous studies on how the middle class, due to their socio-political origins inside, or institutional ties to, the party-state, will share different degrees of fundamental interests with the party-state and hence be unlikely to challenge the status quo (Chen and Dickson 2010; Tomba 2009; Tang and Unger 2013; Goodman and Chen 2013; Chen and Lu 2011). However, such exogenous factors alone cannot explain why the same sample of middle-class respondents presented such paradoxical attitudes, as the key lies not only in the features of their “middle-ness,” but also in how they rationalise different political choices in different contexts.

Indeed, the point of this article is to show that a mismatch exists between how the same group of middle-class respondents approached abstract democratic concepts, real-life political events, and the concept of reform. Although their political attitudes are by no means necessarily representative of the Chinese middle class as a whole, we should pay attention to their varying levels of cognitive reasoning, which shed light on why the middle class have exhibited paradoxical political attitudes in previous literature. It is precisely because they, as middle-class individuals, do not share any coherent “middle-class characteristics” in their political attitudes that makes this subject worth exploring. Most importantly, it should be noted that the same group of middle-class respondents are capable both of recognising the merit of certain political concepts that are crucial to a democratic regime, such as civil society and diverse electoral practices, and of feeling justified in advising against the real-life implementation of these concepts. This level of pragmatic caution is not simply due to political apathy or fear of state retaliation. In fact, it is precisely because they have (or believe themselves to have) a good understanding of China's socio-political environment at large that they have dissociated themselves from the “rash” elements advocating reform. Although some respondents also warned that the middle class are not afraid of voicing grievances should the need arise, China's most recent change in leadership and its call for internal reform seems to have appeased them enough so that they do not think any level of social instability is likely. Thus, it appears that most middle-class respondents equate democratic reform with instability, and vice versa, regardless of how they perceive democratic concepts in general. They are opposed to the idea of reform because of its potential destabilising effect and they could only foresee themselves advocating drastic political change in the face of extreme socio-economic grievances.

The reason behind this line of thinking could be largely historical. The idea of people-based governance, dating back to Mencius (ca. 300 BCE) and beyond, has always measured the legitimacy of the Chinese state by its ability to satisfy the people's needs. Historically, this has been done with no political input from the people: it is a trust-based relationship in that the people grant the state legitimacy as long as they feel the state is making decisions that are beneficial to them. Indeed, if the state fails to meet the people's expectations, it is seen to have lost the Mandate of Heaven; hence, dissent becomes legitimate. However, unless the people's livelihoods are genuinely threatened, the risks and costs of change are perceived to be too high to justify its potential rewards. Those who supported active participation in the PX incident, for example, typically argued that the project had “crossed a line” by endangering the health of the local residents; thus, the protest was seen as an appeal to the local government to hear the protestors’ grievances. Those who did not participate, by contrast, typically did not subscribe to the idea that the project would have any severe adverse effect on their lives, thus they did not feel the need to “take to the streets” in protest. This line of thinking applies only to real-life situations, not to support for democratic concepts, as the latter requires no personal input or commitment from the respondent. This also explains why there is a mismatch between the middle-class respondents’ high levels of support for concepts and their caution about participation. As such, the key aspect of the state– society relationship in China is still a social “gentleman's agreement” (Ling and Shih 1998), whereby the state is expected to heed the needs of the people, instead of a social contract where the relationship is characterised by an exchange of rights and duties. Indeed, soon after the government announced that the PX project had been put on hold indefinitely, the protests were over. Consequently, decisions about political mobilisation and participation are made on a case-by-case basis (as opposed to group consciousness), as the respondents remain watchful over the state's performance and judgement in tending to their needs.

Therefore, the data presented in this article show that to imagine the Chinese middle class as holding the role of either social stabilisers or agents of change is to portray them as more proactive than they are. The respondents in this article are perhaps better labelled as “passive observers”: watchful at best, but extremely cautious about instigating change. Indeed, with the knowledge that they have about the current system, they are largely disinclined to change the status quo. This is, however, not as simple as stating that these middle-class people are conservative in their political outlook. It appears that they do not hold a particularly positive or negative view of the status quo, but rather see it “as it is.” For them, it is most important that they understand the system and play well within it. So far, they still count among the beneficiaries of the system, so they are unwilling to change the rules or reshuffle society in a way that might harm their interests. However, this is not quite the class interest embedded in the party-state that previous literature has explored. Rather, it is a universal interest in socio-political stability and their belief that the state remains the actor most capable of upholding such stability. There is an inherent distrust of the “plebe,” the rash elements within society that they distance themselves from. They are against reforms instigated by the public, which they see as inherently destabilising. By contrast, they welcome top-down reformist agendas arising from within the party-state. This trust in the central government, but not in the general populace, corresponds with a large number of surveys on this topic which find that levels of institutional trust among the Chinese are higher than those of social trust (Ingelhart et al. 2010; Yang and Tang 2010; Rao et al. 2013), and with a multitude of studies that have examined the mismatch between political dissatisfaction and regime support among the Chinese population (Chen et al. 1997; Ren 2009; Wang 2005). The middle-class respondents of this survey, certainly, are less likely to instigate change.

Furthermore, this study has seen no evidence to suggest that these members of the so-called “middle class” view themselves as a separate political entity, one that is on equal or similar terms with the state. Their support for democratic notions is within the complex set of expectations they have for the state and it should be analysed within those contexts. There is no desire or need for these members of the middle class to take proactive action, as they do not (some might argue, dare not) situate themselves opposite the state. However, neither will they necessarily always position themselves alongside the state. It might be presumptuous to assume that these middle-class respondents support the state due to their vested interests, because their expectations may very well change. So far, their interest in the state is largely economic, as they are the beneficiaries of the reforms of the last few decades. Such economic expectations are fundamentally reliant upon socio-political stability, which is their number-one priority. Hence, the paradox between their beliefs and their behaviour: liberal in attitude surveys, conservative in political action, they are very adept at differentiating what “should be” (values of democracy), what “could be” (limited state-led reform to minimise the risk of social instability), and what “is” (the current reform efforts by the new leadership). In a sense, the state has met their expectations, and the legitimacy of the current socio-economic order can be established. Should those expectations evolve, the state must adapt in order to respond to them. So far, the path being taken by the new leadership is deemed as promising: in tightening controls on the bloated civil bureaucracy, it has begun to address one of the key issues of inequity in society, which has been met with approval. Yet, to address one side of injustice is inevitably to harm certain vested interest groups; so, how the state chooses to balance these expectations could be the key to ensuring future regime stability.