Our research indicates that RNR-related access barriers on Chinese social apps have had a profound impact on online communities. In an effort to assess the extraterritorial impact of China’s RNR system and app store censorship, we examined the download patterns of Chinese social media apps for which we identified access barriers. This analysis relied on a dataset maintained by app store intelligence provider Appmagic, covering monthly download data for apps developed by China-based companies from 2015-2023 (see Appendix 1 - Data analysis for further details). The data provides a quantitative proxy for the direct or indirect global impact of China’s RNR system as a censorship tool.

The post-2017 drop in Chinese app downloads by transnational users

Download trajectories revealed in our analysis indicate that China’s great ideological security firewall has effectively limited exchange channels between global communities and PRC-based users. Across a key period of our analysis, we estimate downloads of affected apps declined by an astonishing 82%, amounting to a reduction of 23.5 million downloads. In 2017, the identified apps were downloaded 32 million times — yet by 2023 this figure had dropped to just 5.7 million downloads worldwide (see Table 2, below).

As discussed in Section 4.2 above, major legislative changes in China took effect in 2017 to reinforce the RNR obligations of Chinese internet companies. The subsequent impact and implementation of these regulations coincided with a considerable decline in overseas downloads for apps for which we identified access barriers. While this decline in downloads is significant, it is important to note that other social apps — for which we did not identify access barriers — declined by 40% in that period. This general downward trend indicates that reasons other than RNR-related access barriers contributed to the decline. Therefore, the above mentioned 23.5 million should be lowered to about 14.1 million, for a more accurate estimate for the reduction effect.

Nonetheless, in virtually all regions, monthly downloads for affected apps dropped decisively in a matter of two to three years following the implementation of the RNR requirements. The impact was particularly significant for apps which experienced both types of access barriers, with a drop of 98% (from more than 27.8 million downloads in 2017 to roughly 484,000 in 2023). In North America, apps affected by both phone number barriers and app store censorship dropped from 9 million annual downloads in 2017 to below 1 million in 2020. In the Asia-Pacific region, Chinese social apps peaked in 2017 with 12 million annual downloads, but have since dropped to just below 3 million. Similar trends also occurred in the European Union and UK, Eastern Europe and Central Asia, and the MENA region — a trend that is likely poised to continue as more Chinese apps incorporate domestic RNR requirements for transnational users.

Amidst this considerable decline, our research still identified some recorded downloads for the affected apps. Whether these downloads are the product of successful censorship circumvention by individuals, or the result of the app intelligence provider’s data gathering methods, we cannot determine. The remaining downloads could, however, present a gap in China’s ideological security firewall that censors may subsequently attempt to close. Indeed, based on three-year averages of downloads for the affected apps (2020-2023), we project that more than 5.5 million downloads of Chinese social apps could be further restricted by hard access barriers in 2024. This includes 4.2 million app downloads affected by phone number barriers, an additional 230,000 affected by app store censorship, and 1 million that face both.

Effects on Chinese diaspora communities

Our findings reveal that access barriers disproportionately affect diaspora networks. To assess these impacts, we selected all Chinese social media apps that showed downloads in 2023 (resulting in a group of 48 apps across the 58 countries). We then plotted a relative measure of their success against the size of Chinese diasporas in the given country (see Figure 8, below). The relative measure (vertical axis) refers to the ratio of average monthly downloads of an app per 10,000 inhabitants in a given country. This ratio was calculated based on download data from Appmagic and global population data from the World Bank. In turn, we plotted that ratio against diaspora numbers (horizontal axis) based on conservative estimates drawn from the OECD and the UN (see Appendix 1 - Global population data and Diaspora data for further documentation).

Each dot in Table 4 represent download values for an individual app in one particular country in 2023. If a dot is plotted further to the top right of the graph, the app it represents was frequently downloaded in a country that hosts a large Chinese diaspora. As suggested by the positive inclination of all trend lines, the popularity of Chinese social apps increases in countries with larger Chinese diaspora communities.

Surprisingly, however, access barriers were found to be more common in countries with larger diaspora communities. App store barriers (blue dots), including six country-app pairs such as for Kuaishou in Singapore, were especially present in countries with large diaspora communities. Conversely, apps not affected by access barriers (cyan dots), such as QQ in Pakistan, were more common in countries with smaller diaspora numbers. This is further demonstrated by the fact that the three access barrier trend lines (blue, green, and red) on the scatter plot predominantly lie above the trend line of apps facing no barriers (cyan). A statistical test (one-tailZ-test) confirmed a statistically significant difference, indicating that larger diaspora countries encounter more challenges regarding app accessibility (see Appendix 1 - Statistical testing for further documentation).

Although these results may initially appear to be contradictory, they allow for an interesting interpretation. The Chinese diaspora, while often facing access barriers, managed to bypass government-induced censorship and continued downloading barred social apps.

In the absence of the availability of U.S. social media platforms in China, transnational networks such asthe Chinese diaspora are particularly vulnerable to access barriers on Chinese apps. Chinese-designed digital platforms often constitute the only means through which these populations can interact with online communities back in China. As a result, survey respondents overwhelmingly agreed that access barriers had a significant limiting impact, particularly regarding access to news and information (47%). Limitations were also reported for educational opportunities (29%), communication with professional networks (28%), and maintaining connections with friends and families at home (23%).

On a broader note, these findings highlight how the Chinese government’s ongoing attempt to guide its netizens’ cognition of the world has resulted in the widespread implementation of RNR policies, which have created to detrimental access barriers around social media apps with profound impact on downloads, especially within the Chinese diaspora. We therefore find that the CCP’s attempt to secure the grey zone internet — and, in turn, China’s domestic online discourse — from politically “dangerous” foreign ideas has achieved considerable successes.