How do public opinion and nationalist sentiment affect the foreign policy of China and other non-democratic states? I argue that by allowing nationalist protests against foreign states, non-democratic leaders can use domestic politics for international gain. In China, anti-Japanese protests were tolerated in 1985 and 2005 but banned in 1990 and 1996. Anti-American protests were permitted in 1999 and 2003 but repressed in 2001. Similar patterns of repression and facilitation are readily apparent in Egypt, Iran, Syria, and other non-democratic regimes. Why, when, and how do authoritarian governments give their citizens a green, yellow, or red light to protest against foreign targets?

I develop a theory of anti-foreign protest that suggests that Chinese and other authoritarian leaders have incentives to allow anti-foreign protests in order to gain diplomatic bargaining leverage. A large body of literature has argued that domestic constraints provide advantages in international negotiations. In particular, democratically-elected leaders often state that their hands are tied by constituents or parliamentarians who will punish them at the polls if they back down during negotiations. These potential “audience costs” represent a bargaining tool in international negotiations. Although authoritarian leaders are not constrained by the same electoral institutions, I argue that anti-foreign protests provide an alternative mechanism by which domestic politics can be leveraged in international bargaining. Because anti-foreign protests may turn against the government, allowing such protest makes it costly for the government to make diplomatic concessions and demonstrates resolve in international bargaining.

To evaluate the theory and its implications, I draw upon quantitative and qualitative data gathered over 12 months of field research in China, Hong Kong, and Japan, including more than 100 interviews with government officials, nationalist activists, protest leaders and participants, and foreign policy experts. I also make use of Chinese government documents, press reports, and internet archives. Three case studies, a comparison of anti-Japanese protest in Hong Kong and mainland China, and computerized content analysis of official and commercial Chinese media provide rich support for the theory.