Chinese State Media Seeks to Influence International Perceptions of Hong Kong Protests

Posted: 22nd August 2019
By: Insikt Group®


Click here to download the complete analysis as a PDF.

Recorded Future analyzed data from several Western social media platforms from June 1, 2019 through August 14, 2019 to determine how the Chinese state exploits social media to influence Western public perceptions of the Hong Kong protests. This report details those techniques and campaigns using data acquired from the Recorded Future® Platform, social media sites, and other OSINT techniques. This report will be of most value to government departments, geopolitical scholars and researchers, and all users of social media.

Executive Summary

In March 2019, we published research on Chinese English-language social media influence operations and revealed that they are seeded by state-run media, which overwhelmingly presents a positive, benign, and cooperative image of China. We observed the Hong Kong protests this summer to determine whether Chinese state-run influencers would employ these same techniques when confronted with a domestic crisis.

Based on a retrospective analysis of Chinese state-run English-language social media influence accounts beginning on June 1, 2019, we found:

  • Chinese state-run English-language social media accounts largely ignored the protests in Hong Kong and discussion of the extradition bill until June 9.
  • Sentiment and content analysis reveals that Chinese state-run influence accounts simultaneously condemned the protests and expressed support for the government of Hong Kong. Much of the negative rhetoric and messaging employed by these accounts has been broadly outweighed by posts of positive and supportive messages for the Hong Kong government.
  • The volume of posts from state-run influence accounts increased at critical points in the progression of the protest movement, We assess that this was intended to influence and reframe Western perceptions of the protests.
  • State-run influence accounts referenced the People’s Liberation Army (PLA) in posts about Hong Kong more than any other entity, person, or organization aside from Chief Executive Carrie Lam. We believe this was probably an attempt to seed the social media environment with reasons to justify a possible future mainland intervention in Hong Kong, but it is not an indicator or warning for PLA action.


Insikt Group conducted a retrospective analysis of all of the accessible English-language social media posts from accounts run by Xinhua, the People’s Daily, and four other Chinese state-run media organizations geared toward a foreign audience on Western social media platforms. We limited the data set to these state-run media organizations’ posts and reposts about Hong Kong beginning on June 1, 2019 and running through August 14, 2019. Although the genesis for these protests in Hong Kong — the expanded extradition bill — was introduced in April 2019, the large-scale protests began to increase in size and frequency in early June. We limited our retrospective analysis to June to capture Chinese state-run media sentiment and influence attempts as the Hong Kong protests gained momentum and international exposure.


Our analysis indicates that Chinese state-run English-language social media accounts largely ignored the protests in Hong Kong and discussion of the extradition bill until June 9.

Recorded Future timeline highlighting the progression of the Chinese state-run media’s portrayal of the Hong Kong protests in English-language social media sites.

On June 9 and 10, several state-run media accounts posted stories about the violence of the Hong Kong protestors toward police and characterized the protests as “riots” and “public order offenses.” Other posts from the accounts highlighted Hong Kong Chief Executive Carrie Lam’s press conference and included quotations from Lam calling for “rational and peaceful” discussions on the extradition law and absolving the central government in Beijing from interference in the territory.

Over the course of the next two months, the volume of posts from state-run accounts increased at pivotal times, which we assess was intended to influence and reframe Western perception of the protestors’ actions. The volume of state-run account posts on Hong Kong increased at several critical points.

Thus far, Chinese state-run media account activity surrounding the Hong Kong protests peaked on August 14, with descriptions of protestor attacks on two visiting mainland Chinese citizens, and characterizations of the protests as “radical,” “terrorism,” and akin to a ”color revolution.”


Among our data set, several prominent messages emerged from state-run media’s characterization of the protests:

  1. Hong Kong protestors are undermining the rule of law.
  2. The central government is not responsible for extradition law, but is supportive of Hong Kong leadership and attempts to restore order.
  3. The Hong Kong police are victims of the violence and are reacting proportionately.
  4. Western powers are interfering in Hong Kong, and Western media only shows one side of the violence.
  5. Protestors are hurting Hong Kong’s economy and negative economic consequences are already being felt.

Further, during this time period, state-run media accounts referenced one organization more than any other — the PLA. Our analysis demonstrates that when state-run media accounts posted about Hong Kong, the most frequent organization also referenced in those posts was not the Legislative Council or the Police Force, but the PLA. State-run media referenced the PLA in posts about Hong Kong more than any other entity, person, or organization, aside from Chief Executive Carrie Lam.

PLA-inclusive posts, or posts that reference both Hong Kong and the PLA, began on June 24, when state-run media accounts disseminated information about the PLA Hong Kong Garrison opening to visitors for three days later in the month. After subsiding for a few weeks, PLA-inclusive posts then restarted in mid-July.

Examples of PLA-inclusive messages posted by state-run influence accounts.

We assess that by mid-July, state-run media had begun to seed the social media environment with reasons to justify a possible future mainland intervention in Hong Kong. While these PLA-inclusive posts and co-occurrences are significant, we do not believe this is an indication or warning of possible PLA intervention in Hong Kong.

Sentiment and Content Analysis

Our sentiment and content analysis of these influence accounts’ social media posts since June 1 revealed a complex narrative and delicate balancing act. Our sentiment analysis indicated that over the course of this time period, state-run social media influence accounts posted solidly positive content regarding Hong Kong. On average, the sentiment score for content posted by these accounts was 0.273, or largely positive on a scale where anything greater than 0.05 is considered positive (more on sentiment analysis in the paragraphs below).

At the same time, as mentioned above, Chinese state-run media has at times employed negative and inflammatory language to characterize the actions of the protestors. While this language is certainly reflected in our social media data set, this negative rhetoric and messaging has been broadly outweighed by posts of positive and supportive messages for the Hong Kong government. Examples of this type of messaging are shown in the images below.

Series of posts from state-run social media accounts demonstrating the balance between criticism of the protestors’ actions and positive support for the Hong Kong government.

Further, during this time frame, these accounts condemned the actions of the protestors — not the fact of protest in and of itself. We assess that this condemnation of violent actions, combined with overt support for the Hong Kong government, economy, and law enforcement is likely part of a strategy to positively frame the central government in Beijing’s role in these events.

For the six months prior to June 9, state-run social media accounts posted content about Hong Kong that was solidly positive, with an average sentiment score of 0.371. Typical scoring thresholds used in sentiment analysis are:

  • Positive: Greater than 0.05
  • Neutral: From -0.05 to 0.05
  • Negative: Less than -0.05

A score of .371 reflects strongly positive sentiment. On average, Chinese state-run influence accounts projected positive sentiment regarding Hong Kong to platform users prior to June 9, 2019. As demonstrated in prior research on Chinese state-run social media influence campaigns, Chinese state-run influence accounts generally project positive sentiment to platform users, which is consistent with the strategic goal of portraying China’s development and rise as positive and beneficial for the global community.

We assess that this strategy of employing positive messages remains consistent in state-run media attempts to influence Western perceptions of the Hong Kong protests. While the average sentiment numbers after June 10 declined slightly, state-run media accounts still messaged strongly positive sentiment about Hong Kong for the subsequent two months, returning a score of 0.273.

Another factor likely affecting the generally positive sentiment on Hong Kong was the death of former Chinese Premier Li Peng. As we will explain in more depth below, Li Peng, who passed away on July 22, played a pivotal role in bringing Hong Kong back under mainland administration, under the “one country, two systems” policy. State-run influence accounts were broadly laudatory of Li Peng and highlighted his role in Hong Kong’s return. These messages and this sentiment likely played a small part in raising the positive sentiment score for these accounts.

Sentiment analysis is simply one technique for general analysis of social media posts. There were certainly negative messages and sentiments among the posts by state-run influence accounts. However, the overall trend during this time frame indicates that these accounts on average promoted positive or supportive sentiments over violent or negative ones in an attempt to frame the discussions around the Hong Kong protests.

The Death of Li Peng

Somewhat awkwardly, the death of former PRC Premier Li Peng on July 22 complicated state-run media’s attempts to characterize the Hong Kong protests because of Li’s role in the 1989 Tiananmen Square crackdown. Li Peng is widely known outside of China as the “Butcher of Beijing” for his role undermining more moderate Party leadership and calling for “decisive measures to put down this counter-revolutionary riot” the day before military forces violently cleared the square.

From July 22 through July 29, state-run media posted messages extolling the virtues of Li Peng and his role in implementing the “one country, two systems” policy under which Hong Kong was rejoined with the mainland. The below image is emblematic of the posts from Chinese state-run media at the time and included comments from social media users linking Li Peng to the Tiananmen Square massacre and drawing parallels to the Hong Kong protests.

Screenshot of a post from the Global Times about the life and legacy of Li Peng.


This research examines Chinese state-run influence attempts during a specifically critical and difficult period for the central government in Beijing. It is an effort to characterize how Chinese state-run media is attempting to influence Western public perceptions of one timely issue — the protests in Hong Kong. We do not believe these messages provide indicators or warning of potential central government intervention in Hong Kong, but do reveal the techniques Beijing continues to deploy to try and influence Western publics.