From Andrew Schwartz/ The Center for Strategic and International Studies (CSIS)
SOURCE: “Public Opinion in China: A Liberal Silent Majority?” by CSIS’s Ilaria Mazzocco and Scott Kennedy.
Big Data China is a new collaboration between the CSIS Trustee Chair in Chinese Business and Economics and the Stanford Center on China’s Economy and Institutions (SCCEI) whose goal is to bridge the gap between cutting-edge quantitative academic research and the Washington policy community. Through regular multimedia features, written analysis, and public events, Big Data China identifies, analyzes, and introduces top scholarly research on China and then teases out the most relevant findings for today’s policy issues.
This first feature analyzes data which sheds new light on the views of Chinese citizens, challenging some mainstream scholarship on the topic. Survey data shows that Chinese citizens hold diverse views on many policy topics and do not always support government choices.
Stay tuned for regular new features and a dedicated microsite. We welcome your feedback to ensure that Big Data China is a valued resource for the policy community.
-Scott Kennedy, CSIS, and Scott Rozelle, SCCEI
Media www.rajawalisiber.com – Increasing centralization of power in China and reduced in-person exchanges with the West have driven the perception that only one person’s views in China matters: Xi Jinping’s. In addition to not having much power, most assume that Chinese citizens’ views are largely shaped by the Chinese Communist Party (CCP) propaganda machine. Most in Washington believe that to the extent that Chinese citizens have independent views, they would not dare to share them because of the dangers of doing so.
This simplistic view of Chinese public opinion is off base. People in China have diverse and well-formed views on a wide range of public policy issues. Not all citizens are supportive of current government policies, nor do all their views reflect state propaganda. And, despite the risks, they are willing to share their opinions. These are the clear findings that emerge from quantitative research by Stanford professors Jennifer Pan and Yiqing Xu.
Survey data collected through a variety of channels and methodologies by Pan and Xu over several years show that Chinese urban residents are more liberal than expected and more liberal than the official positions of their government. Moreover, the political views of respondents remained relatively stable over time and were correlated across issues in ways comparable to those in democratic countries.
Pan and Xu’s work sheds new light on the views of Chinese citizens, challenging some mainstream scholarship on the topic. They find diverse views on many policy topics, suggesting that Chinese citizens do not always support government choices.
The implications of their scholarly research for the policy community are profound. American policymakers assume that Xi Jinping and the CCP face no challenges from domestic public opinion except perhaps from hypernationalist groups. But public opinion can diverge from the party line—and it is more diverse and liberal than one might expect. There are plenty of nationalists, but there is also a silent majority in favor of economic reform and political liberalism.
Better understanding the role of public opinion and ideological views in China could provide powerful insights into some of the drivers and constraints of Chinese policymaking. In particular, it can shed light on levels of support for the regime, especially from the middle class and entrepreneurs, which are relatively new social groups in post-Mao China.
Many studies based on public opinion surveys, focus groups, and interviews in China have found that support for the regime and satisfaction is relatively high, especially among the middle class and entrepreneurs. For these analysts, this alignment of views between the public and the party-state has contributed to the regime’s stability. They also indicate that those who had benefitted economically from government policies and were closer to the center of power were more likely to support the regime and oppose political liberalism (see Box 2, below).
By contrast, Pan and Xu’s research shows that, on average, wealthier and more educated Chinese are more, not less likely, to hold politically liberal, pro-market, and non-nationalistic views. This is particularly significant at a time when the government is taking an increasingly illiberal, statist, and nationalistic turn. While not a sign that China is on its way to democratization, the data indicate that the party-state must contend with well-formed ideological views among its citizens that diverge with its own and is pursuing policies that face substantial, even if quiet, public opposition.
The relationship between nationalism, public protests, and foreign policy also holds important implications for U.S. policymakers. There are varying interpretations of Chinese nationalism (see Box 2, below): some believe that protests are cultivated by the government, that policymakers are constrained by protesters, or both. A key conclusion is that an influential portion of the population holds nationalist views that can easily be inflamed.
Pan and Xu argue that plenty of Chinese are very patriotic but very few are supportive of war in pursuit of those goals. This is especially true for wealthier, more educated respondents. This finding echoes the results of a survey from two decades ago, which showed that China’s emerging middle class was far more internationalist and opposed to militarism than other social groups.
There are several insights from Pan and Xu’s research that deserve Washington’s attention and that are highlighted below. These include the surprising stability of Chinese public opinion; the unexpected breadth of views on issues related to politics, the economy, and nationalism; and the systematic variation in opinion that suggests that the public will grow more liberal as the country modernizes and interacts more deeply with the rest of the world. These insights should affect how Washington develops and communicates America’s China policy.
Pan and Xu have not only shown that there are more liberal and coordinated tendencies in Chinese society than is generally believed on topics concerning politics, economics, and foreign policy. They also have proven that there is a systematic pattern in how respondents answer that holds over time.
Using surveys conducted from 2012 and 2014 and separately in 2018 and 2019 Pan and Xu show that Chinese respondents have coherent policy preferences that are bundled in predictable ways. For example, those who hold politically liberal views are more likely to also support free markets and oppose nationalistic foreign policy. Those who support authoritarian political institutions are instead more likely to support state intervention in the economy and a nationalistic foreign policy (Figure 1).
China’s authoritarian political system makes asking the Chinese public their political views difficult but far from impossible. To address this challenge, Pan and Xu ask multiple policy-specific questions rather than for general ratings of the regime or the political system. This approach reduces the likelihood respondents will engage in self-censorship or be put at risk by participating in surveys (for more on doing surveys in China, see Box 1, below).
Chinese citizens have a range of opinions about individual rights and political freedoms that do not always match existing policies or state propaganda. For example, a majority of respondents in the 2018 and 2019 surveys believe that the government should not interfere with whether people have children or how many children they have (Figure 2). This is clearly a negative judgment on the decades-long “one-child policy,” which restricted most families from having more than one child. It also holds implications for the current reversal of the policy and the recent indications that the government is pressing families to have more children, suggesting it will be harder than one would expect to get the public to comply.
Negative views on government interference in family planning are less surprising given the relative increase in personal freedoms in China since the end of the Mao era and the gradual relaxation of such policies. But repressing freedom of speech and free assembly and demonstrations are at the heart of the CCP’s policies, and the space for free expression and criticism of the government has only narrowed in recent years.
Nonetheless, a clear majority of respondents, almost 60 percent in surveys conducted in 2018 and 2019, supported allowing people to express their views on government policies, whether positive or negative (see Figure 3). Moreover, a large minority opposed limitations on free assembly and demonstrations (see Figure 4).
Another study by Pan and Xu challenges conventional wisdom about the Chinese public even more. In an experiment, researchers showed respondents randomly paired bundles of policies and asked them to select their preferred policy. They found that most respondents preferred allowing free speech even if it were to lead to social instability. This is striking as it clearly indicates that most respondents valued free speech over social stability or that respondents do not believe that such freedoms necessarily undermine stability.
The finding is surprising because of the conventional view that Chinese individuals—in both officialdom and the public—are highly averse to public conflict and anything that could potentially create “chaos.” Previous research indicated that the average Chinese citizen was highly risk averse and placed much more value on social stability than on political freedom. But if large portions of the population are not concerned that free speech and public assembly present threats to social stability or value these personal freedoms over stability, then this raises questions over the popularity of current repressive policies. The findings show there to be less acceptance of current trends in governance than previously thought and more unspoken support for protests.
The Xi Jinping administration has been strengthening authoritarian control, but it is also pursuing a far more interventionist approach toward the economy than its predecessors, in the form of “state capitalism.” This involves intensive industrial policies across a range of sectors, a clampdown on the private sector, and greater support for state-owned enterprises (SOEs). This trend has created increasing tension between China and its trading partners, but there has been little analysis of where the Chinese public sits on these issues—until now. Pan and Xu’s research finds that even within China, views on the economy are not entirely in line with those of the Xi regime.
To determine the public’s views about the appropriate role of the private sector, Pan and Xu asked Chinese citizens in two rounds of surveys in 2018 and 2019 several questions about the role of the private sector and the government in the economy.
In one question, survey participants were asked whether it would be appropriate for the private sector to provide higher quality healthcare services for those who are willing to pay higher prices. This is controversial given that China has a public healthcare system in which the vast majority of hospitals are state owned and operated (see Figure 5). Pan and Xu found that in both surveys, the majority of respondents supported allowing private healthcare services even if it meant that those willing to pay more received better care.
Beyond the specific level of acceptance of private investment in healthcare, the responses also show that the public is willing to accept a certain level of inequality in what the state treats as a public good. This is important because inequality is increasingly politicized in China and is the main target of the Xi administration’s “common prosperity” program.
Living standards in China have increased significantly over the past four decades since the beginning of the Reform and Opening Era. But inequality has shot up as well, raising concerns among policymakers who are now touting a policy of “common prosperity.” But Pan and Xu’s research suggests that at least in areas such as healthcare, discontent with wealth inequality may be less widespread than assumed. This is consistent with previous research. In fact, there may be some ready to pay a premium for better services who will oppose efforts to reduce private provision of public goods and the “common prosperity” program.
Land ownership is an even more controversial topic since it is owned by the state and historically has been the object of widespread collectivization policies. Land policy also has implications for equity given that it has provided a safety net for many rural residents.
The survey shows that there are split views on land ownership, with a slim majority of respondents opposing the proposition that “individuals should be allowed to own, buy, and sell land” in both iterations of the survey (see Figure 6). This is not surprising considering how politically sensitive the topic is, but it is notable that in both surveys, a third of respondents stated that they supported individuals being allowed to own and sell land. There is a sizeable and stable minority that supports expanding economic liberties and would oppose growing statist interference in the economy.
There is more evidence that a large number of Chinese do not approve of the government’s interventionism. In an experiment where Pan and Xu showed respondents randomly paired bundles of policies and asked them to select their preferred policy, the preferred position was to reduce support for state-owned enterprises and reduce taxes for private enterprises. In other words, respondents want a level playing field where market performance rather than public ownership determines the success of a company.
Support for economic liberalism only goes so far when it comes to trade, however. For example, a large majority of respondents support tariffs on foreign goods to protect domestic industries (see Figure 7).
The exact relationship between Chinese foreign policy and nationalism, including its most visible incarnation in street protests against Japan and the United States, has been the object of several studies (see Box 2, below). Is the government orchestrating protests through its jingoistic propaganda, or are ultra-nationalists in China shaping the country’s foreign policy? Pan and Xu’s surveys from 2018 and 2019 show that while patriotic feelings are widespread, they stop short of supporting actual war.
For example, most respondents in both versions of the survey supported the proposition that the government should “safeguard national sovereignty and territorial integrity as much as possible through diplomatic and economic means to avoid military conflicts” (see Figure 8). This suggests that citizens do not see war as a desirable tool to protect national sovereignty and would oppose policies openly provoking it.
Public aversion to military conflict is further supported by a survey experiment where participants selected their preferred policy from a randomly paired bundle. The results show that most respondents supported maintaining or increasing China’s military presence in the South China Sea. Support for military expansion, however, declined significantly when it came at the expense of war.
This suggests that while the population is generally in agreement with nationalist rhetoric, support for an actual war is low. This is even more striking given that China has not been involved in any protracted military conflict in well over a generation, something that can drive public opinion against militarism. Public opinion could in effect constrain the top leadership’s appetite for war even beyond the South China Sea.
What complicates the picture is that despite ambivalence toward war, nationalism is popular. In the survey experiment, only one choice was more unpopular than increasing military presence at the cost of war: reducing China’s military presence in the South China Sea to maintain friendly relations.
On questions not pertaining to war, Pan and Xu found widespread nationalistic attitudes. For example, most survey respondents supported forbidding the entrance of foreign journalists who regularly publish negative reports on China (see Figure 9). A clear majority also opposed allowing dual citizenship, which is currently not permitted for Chinese nationals. Opposing dual citizenship indicates a strict understanding of nationality since allowing it would imply accepting a dual national identity and more openness to international ties (see Figure 10).
This suggests that, as is often the case, nationalism is multifaceted. Popular nationalism is reinforced by and fuels government policies, yet the surveys show China’s populace has not given their unequivocal support for engaging in war to defend official Chinese claims, limiting how much the government can act on its military threats.
Xu and Pan’s research also shows that ideology in China is correlated with income and education in ways that remain stable over time. In surveys conducted between 2012 and 2019, Pan and Xu found that respondents with higher incomes and higher levels of education were more likely than respondents with lower income levels and lower levels of education to hold views that were liberal, pro-market, and less nationalistic.
Wealthier respondents were more likely to support politically liberal and pro-market policies (see Figures 11 and 12). Those who had attained higher-level degrees (college and post-graduate degree level) were also less in favor of militaristic policies and nationalism compared to respondents with lower levels of education.
The evidence shown in Figures 11 and 12 has some important implications. It indicates that the country’s elites, composed of the most educated and wealthiest members of the population, are not supportive of more authoritarian, statist, and nationalistic policies. Public opinion in an authoritarian country such as China has a less direct impact on policymaking than in democratic countries, but elite support is important to regime survival. Moreover, the voice of the middle class is generally more influential in shaping policy compared to that of lower-class citizens.
Pan and Xu’s findings that the wealthier and more educated segments of society are relatively liberal also provide an update to some of the existing assumptions on Chinese middle-class attitudes. Previous research had indicated that Chinese elites are more supportive of the regime than other segments of the population thanks to their links to the leadership and because they have benefitted from the party-state’s policies.
Analysts of Chinese public opinion have long focused on a key demographic, university students, because they are likely to be future elites with a great deal of influence over the country’s future. An important subset is those who study abroad. There were over 370,000 Chinese students studying at American universities in the 2019-2020 school year when COVID-19 hit. While educational exchanges were seen for decades as a tool to promote American values and friendliness between the two countries, they are increasingly controversial. Recent concerns include the theft of intellectual property and allegations that some Chinese student organizations act on the behest of the party threatening freedom of speech in American universities.
Pan and Xu conducted three surveys of Chinese students enrolled in three top Chinese universities and Chinese students enrolled in 60 U.S. institutions to compare their ideologies. The data show that Chinese students who study in the United States are on average more politically liberal, pro-market, and less nationalistic than those studying at top universities in China (see Figures 13-15). The finding has held constant even in 2020, despite political instability in the United States, a rise in racist attacks toward Asian people living in the United States and the Asian American and Pacific Islander community, and disruption due to the pandemic.
A possible explanation is that more liberal, pro-market, and less nationalistic young Chinese people want to study in the United States. Perhaps families that are less supportive of Chinese government policies are also more likely to encourage their children to study in the United States to provide them with different perspectives and opportunities.
The findings have profound policy implications. On average, Chinese students in the United States are less likely to identify with aggressive nationalist ideologies, which means that the overall threat posed by these students to freedom of speech in American universities is limited. This provides more context for the findings in a recent report which showed that while violations of freedom of speech in American universities at the hands of Chinese students did take place, they were not widespread.
The findings from this stream of research have critical implications for American policy toward China. The dominant view in Washington appears to be that public opinion in China is closely aligned with CCP propaganda and is easily manipulated. To the extent there is a small segment of regime critics, their views do not matter because China is authoritarian, and the regime can ignore or crush critical perspectives. The reality, in fact, is that the Chinese populace has a wide range of views that are less vulnerable to state propaganda than most expect. Taking a closer look at current views, it is clear that Chinese citizens have complex yet coherent views that often differ significantly from official government ideological frameworks and policy. In a word, Xi Jinping faces genuine popular domestic political challenges. There is what may be called a “silent liberal majority” that complicates the main thrust of his own approach and is more aligned with the integrationist path the United States and others have hoped for.
The direction taken by the Xi administration on economics and civic liberties, including the “common prosperity” program, may be at odds with the preferences of large portions of the population, especially among more educated and wealthier Chinese citizens. The middle class has been thought to be satisfied with the party’s performance, but the regime’s policy choices could bring a change in that support.
Similarly, nationalism in China has been understood until now to be both a driver and an enabler of the government’s militaristic attitudes toward the South China Sea, Taiwan, and other irredentist disputes. Instead, Pan and Xu’s findings suggest patriotic sentiment should not be equated with a blank check that Chinese authorities could freely cash to go to war. The Chinese population seems to be very conflict averse despite strong levels of nationalism. This attitude could certainly influence how the government makes decisions in a crisis.
These insights can help strengthen America’s China policy. The United States can expect that China’s current leadership, even if more risk tolerant than previous administrations, still faces domestic politics that could constrain both domestic and foreign policies and that such concerns, not just the singular views of Xi, may explain twists and turns in Chinese policies. A better understanding of Chinese public opinion, in all its variations, may help Washington better predict where official Chinese policies will go or what challenges they will face if they are inconsistent with a large swath of public opinion.
In addition, given that some important segments of the Chinese public have more liberal views than the Xi administration, Washington should consider how the framing and defense of its policies would be perceived by these potentially sympathetic ears. Defenses of policies toward the Indo-Pacific region and toward China bilaterally that are entirely antagonistic may lead to a loss of support from certain segments of Chinese society. This in turn may reduce the concerns China’s top leaders have about taking a more conciliatory approach toward the United States and the international community in general. Being attuned to and serving the interests of the American public is the number one job for Washington, but being aware of Chinese popular perspectives can only improve the efficacy of American policy.
Finally, these findings show there are real benefits from maintaining people-to-people exchanges with China. It may be that Chinese students who come to the United States are more predisposed to affinity with America than the average Chinese person, but even so, studying in the United States provides sustenance to those views and strengthens ties that benefit America’s economy and national security. While the United States needs to smartly tackle challenges from Chinese intellectual property rights theft or intelligence operations, those efforts should be done within the awareness that the goals of connectivity have and still can bear fruit.
Box 1 – Methodology: Doing Surveys in China
There are inherent challenges to conducting public opinion surveys in an authoritarian setting such as China. Self-censorship can be a serious problem in determining the real opinions of respondents, especially when it comes to sensitive questions. Pan and Xu address this challenge by asking multiple policy-specific questions for each thematic area (e.g., nationalism, political liberties, economics) that help paint a picture of a person’s belief system. They also sampled different groups over several years and found that the distribution of responses was remarkably consistent.
The exact absolute proportion of people with any one view may not be accurate when scaled to the entire country, but the fact that surveys detect a wide range of opinions shows that respondents are seriously considering the questions and more often than not offering their genuine thoughts. For example, it is striking how many people consistently offer responses that are not aligned with government policy or propaganda. The methodological approach and the breadth of answers, taken together, mean there is high confidence that Pan and Xu’s findings are reliable.
The safety of participants is also a concern because opinions critical of the government may expose them to retaliation. The advantage of an online survey is that respondents’ private data are more protected. They can take the survey in a place and time of their own choosing on the device they prefer, no personally identifying information about them is recorded, and the data are fully encrypted. All surveys have been reviewed by the Institutional Review Board at the authors’ home institutions. The surveys are supported by both Stanford University and the 21st Century China Center at the University of California, San Diego, where both Pan and Xu are affiliated.
For the general population surveys (as opposed to those targeting students), Pan and Xu used quota-sampling to match, to the extent possible, census data for gender, age, education level, and geography in urban China. Because they used an online survey, they were unable to ensure that the sample was fully representative of the country. So, some quotas (e.g., respondents with less than high school education, over the age of 50, and in western regions of China) were not successfully filled.
The survey questions in Figures 2 through 10 and the data from Figures 1, 11, and 12 are based on surveys conducted on two samples fielded in 2018 and 2019, respectively. The combined sample size is 2,329 respondents. To assess the ideology of respondents, the authors asked 14 questions in each policy domain.
Figures 13-15 depict the results from a longitudinal panel survey that was conducted each fall and spring since 2019, or every six months. Pan, Xu, and their co-authors Fan Yingjie and Shao Zijie surveyed Chinese students enrolled in three top Chinese universities and Chinese students enrolled in over 60 U.S. institutions, including public and private institutions of higher education in all regions of the United States (East, Midwest, South, and West).
The data presented in this feature come from three studies conducted by Jennifer Pan and Yiqing Xu, of which two have been published. For more detailed methodological information, see:
Shane Xuan, Jennifer Pan, and Yiqing Xu, “Political Preference in the Shadow of Institutional Backsliding,” Working Paper, 2022.
Box 2 – Further Resources
The following resources provide insight into public opinion and ideology in China:
Shane Xuan, Jennifer Pan, and Yiqing Xu, “Political Preference in the Shadow of Institutional Backsliding,” Working Paper, 2022.
For some recent surveys on Chinese public opinion on a variety of issues including views on the United States, Japan, and attitudes toward the regime, see:
“China From the Ground Up,” The 21st Century China Center, UC San Diego.
The following articles and books address the sources of support for the regime in China and political attitudes of the middle class:
Bruce Dickson, The Party and the People (Princeton, NJ: Princeton University Press, 2021).
Susan Shirk, China: Fragile Superpower (Oxford, New York: Oxford University Press, 2008).
The following works address nationalism and its relationship to foreign policymaking in China:
For more on the ideology and political activities of Chinese students in the United States, see:
Ilaria Mazzocco
Fellow, Trustee Chair in Chinese Business and Economics
Ilaria Mazzocco is a fellow with the Trustee Chair in Chinese Business and Economics at the Center for Strategic and International Studies (CSIS). Prior to joining CSIS, she was a senior research associate at the Paulson Institute, where she led research on Chinese climate and energy policy for Macropolo, the institute’s think tank. She holds a PhD from the Johns Hopkins School of Advanced International Studies (SAIS), where her dissertation investigated Chinese industrial policy by focusing on electric vehicle promotion efforts and the role of local governments. She also holds master’s degrees from Johns Hopkins SAIS and Central European University, as well as a bachelor’s degree from Bard College.
Scott Kennedy
Senior Adviser and Trustee Chair in Chinese Business and Economics
Scott Kennedy is senior adviser and Trustee Chair in Chinese Business and Economics at the Center for Strategic and International Studies (CSIS). A leading authority on Chinese economic policy, Kennedy has been traveling to China for over 30 years. His specific areas of expertise include industrial policy, technology innovation, business lobbying, U.S.-China commercial relations, and global governance. He is the editor of China’s Uneven High-Tech Drive: Implications for the United States (CSIS, February 2020) and the author of The State and the State of the Art on Philanthropy in China (Voluntas, August 2019), China’s Risky Drive into New-Energy Vehicles (CSIS, November 2018), The Fat Tech Dragon: Benchmarking China’s Innovation Drive (CSIS, August 2017), and The Business of Lobbying in China (Harvard University Press, 2005). He has edited three books, including Global Governance and China: The Dragon’s Learning Curve (Routledge, 2018). His articles have appeared in a wide array of policy, popular, and academic venues, including The New York Times, Wall Street Journal, Foreign Affairs, Foreign Policy, and China Quarterly. He is currently writing a report tentatively titled, Beyond Decoupling: Winning the Hi-Tech Competition Against China.
From 2000 to 2014, Kennedy was a professor at Indiana University (IU), where he established the Research Center for Chinese Politics & Business and was the founding academic director of IU’s China Office. Kennedy received his Ph.D. in political science from George Washington University, his M.A. in China Studies from the Johns Hopkins School of Advanced International Studies, and his B.A. from the University of Virginia.
For a full list of Scott Kennedy’s publications, organized by category, click here.
This feature was made possible through the generous support of the Stanford Center on China’s Economy and Institutions (SCCEI). Special thanks goes to Professors Jennifer Pan and Yiqing Xu for sharing their work and time with us, and to the SCEEI team of Scott Rozelle, Matthew Boswell, and Jennifer Choo for the dedication to this collaboration. We also are grateful for the hard work and professionalism of our CSIS colleagues, including the Trustee Chair’s Alyssa Perez and Qin (Maya) Mei.
Special Thanks to:
- Alyssa Perez, Associate Director, Trustee Chair in Chinese Business and Economics
- Qin (Maya) Mei, Research Associate, Trustee Chair in Chinese Business and Economics
- Tucker Harris, Lead Web and Digital Product Designer, iDeas Lab
- Matthew P. Funaiole, Director, iDeas Lab, and Senior Fellow, China Power Project
- Jeeah Lee, Associate Director, Publications
- William Taylor, Senior Designer, iDeas Lab
- José Romero, Associate Developer, iDeas Lab
- Christina Hamm, Associate Web Designer, iDeas Lab
- Cera Baker, Multimedia Producer, iDeas Lab
- Laurel Weibezahn, Multimedia Producer, iDeas Lab
A product of the Andreas C. Dracopoulos iDeas Lab, the in-house digital, multimedia, and design agency at the Center for Strategic and International Studies.