Frontiers of Democracy Research: The Co-existence of Censorship and Responsiveness in Chinese Government

J PanHarvard Academy Graduate Fellow and recent Ash Center Democracy Fellow Jennifer Pan studies the intersection of politics, citizen interaction, and service delivery in Chinese cities. In this post, Richa Mishra captures Pan’s latest work on the nature of the interactions between government officials and citizens—from censorship to organizing to political responsiveness—in a regime without electoral competition. To whom are local Chinese government officials responsive? While we might assume that top down influences dominate, it appears that bottom-up pressure from citizens plays a role as well.

This post is part of an occasional series, Frontiers of Democracy Research, highlighting the research of faculty, fellows, and affiliates of the Ash Center, whose work explores innovations in public participation and other substantive challenges to democratic governance. With the spirit and understanding that the chief driver in democracy is demos, or “We the People,” this series takes discussions out of the realm of academia and into the public realm. This post is based on conversations with Pan in person, over the phone, and by email over the summer of 2014. As with all our posts, we look forward to your comments.


By Richa Mishra

In spirit, one of the key features of democratic governance is the accountability to citizens and our preference that should be afforded by the regular elections of our representatives. In countries with non-democratic political systems, in the absence of electoral competition, how do governments tend to interact with citizens?

Ph.D. candidate and recent Ash Center Democracy Fellow Jennifer Pan, who studies Chinese politics, focuses her research on three broad categories of interactions:

  1. How the government influences public opinion;
  2. How citizens make their demands known to the government; and
  3. How the government responds to citizen preferences.

Influencing Public Opinion

Pan’s research, as part of her doctoral research at Harvard University’s Department of Government, explores how the Chinese government “prunes” or censors the information available to citizens. She also studies how the Chinese government responds to citizens’ demands as articulated through different channels.

In short, the Chinese government’s program to selectively censor the “expressed view” of the Chinese people is unprecedented in both scope and sophistication. It has the ability to control what citizens see in traditional media such as newspapers, television, and radio, as well as new media such as micro-blogging sites, telephone SMS messages and online videos.

The Chinese government in fact allows for a great deal of criticism of the regime, including its policies and officials. Its censorship instead focuses on discussions related to any group-based action that is outside the control of the state, even including pro-government rallies.

Building on this study, the team next employed two new methods to reveal more information about how censorship occurs, including before submissions are posted. First, to establish causality, Pan and her colleagues conducted a large scale experimental study in which they created accounts on numerous social media sites spread throughout the country. Through these account they posted a variety of randomly-assigned types of social media text, and used a network of computers all over the world to detect which types of posts were censored.

Next, the team set up its own social media site in China and contracted with Chinese firms to install the same censoring technologies they use on existing sites. With direct access to these firms’ software, documentation, and even customer service help desk support, Pan and her colleagues reverse engineered the mechanics of social media censorship in China, determining how it all works.

Their results, just published in Science magazine, offer the first rigorous experimental support for Pan’s previous hypothesis that criticism of the state, its leaders, and their policies are routinely published—whereas posts about real world events with collective action potential are censored.


Example of the message that shows up when visiting a censored web page.

Example of the message that shows up when visiting a censored web page.


Receptivity to Citizen Demand

Pan’s team has also found that slightly more than half of Chinese officials were “receptive” to citizens’ demands–meaning that they were willing to incorporate these demands into policy decisions.

In order to gauge local officials’ receptivity to citizens’ demands, Pan and her colleagues conducted a survey experiment among city and provincial level officials. Using a survey technique that protects the anonymity of respondents and helps obtain more honest responses to sensitive questions, they asked the officials whether they would seriously consider citizens’ suggestions in making policy and expenditure decisions related to “minsheng zhengce,” the needs of the people. These needs included education, public health, social welfare, employment, housing, and environmental protection.

While most officials said they were willing to incorporate citizen demands into their decision-making, there were of course exceptions. Pan found that in places where the relationship between citizens and local officials was strained, for instance where protests have occurred in the past, the government was only receptive to citizen demands voiced through “traditional” channels. These channels include representatives of a local people’s congress or residential committees, as opposed to newer channels like social media. The reason why was not clear.

Government Responsiveness and Implications for Democracy

To better understand actual responsiveness to citizens’ demands, a step beyond receptivity, Pan and her team conducted an online experiment in all 2,000+ Chinese counties with a web forum set up and managed by the local governments through which citizens can pose questions and receive responses from local government officials. The research team submitted requests for obtaining social assistance to the government forums, varying the wording of their requests to test hypotheses about what would make local governments more or less responsive.

Upon evaluating the response rate along with the context and content of the responses, they found that governments responded to approximately one third of citizen demands on these forums. Among government responses, the level of helpfulness was mixed. A slight majority of responses provided detailed information tailored to the request, but almost 1 in 4 responses stated that some piece of information was missing, and the rest suggested contacting a different government agency.

When requests included threats to engage in collective action, that is to organize with like-minded fellow citizens to do something about a mutual problem, local governments were 30% more responsive to the demands.

These studies and Pan’s other work examine political interactions ordinarily associated with democracy, such as responsiveness and freedom of information. Studying them in a context without elections, she says, allows her to examine whether and how these interactions are realized in non-democratic systems.

As a recent Democracy Fellow at the Ash Center, Pan participated in a seminar series led by Archon Fung while she completed her doctoral research. She credits the experience with bringing her in contact with political theorists and practitioners, relationships that were instrumental in deepening her understanding of both the theoretical implications of her research as well as its practical, tangible impacts. Pan says, “As a social scientist, it is very easy to remain in the empirical realm, dealing only with data, but the workshop gave me the opportunity to come out of that cave and think more expansively about my work.”

Looking ahead, Pan is interested in continuing to study state-society interactions in authoritarian countries. She believes that there is a tendency to misconceive countries with elections as democracies and countries without as dictatorships.

“Layered on top of this dichotomous outlook is a normative view that reduces to ‘democracy is good and dictatorship is bad.’ But increasingly, the line between what is a democracy and what is a dictatorship is blurred. For example, Russia has had presidential elections since the fall of the U.S.S.R, but is it a democracy today, “Pan asks. “From 1928 to 1982, Mexico Institutional Revolutionary Party (PRI) won every presidential election – was Mexico a democracy before 1982?”

Pan urges us not to rely solely on elections as the bellwether of whether or not a country is democratic. There is a danger, she warns, that we promote elections as a proxy for democracy but that elections may not lead to the democratic values and outcomes we would want to see.


Richa Mishra is a research fellow at the Ash Center for Democratic Governance and Innovation. She has over 15 years of experience in international development and public policy formulation with the UN system, the World Bank and research and academic institutions. Richa has extensive experience in promoting democratic institutions in transitional political systems.

Leave a Comment

Filed under Frontiers of Research, Participation, Representation, Technology

Leave a Reply

Your email address will not be published.