Digital censorship is increasingly common even within democracies, raising questions about political motivations behind it.

The study uses new cross-national data from Google detailing government requests for content removal and user information. It demonstrates that governments censor more when faced with internal dissent or strong economies producing substantial intellectual property (IP).

However, this tendency is moderated by democratic electoral systems: proportional representation reduces censorship demands. The findings suggest a complex relationship where economic factors fuel censorship pressure, but institutional design shapes its implementation.

Google Dataset

The research relies on previously unavailable government requests data provided directly to Google researchers across multiple countries and time periods.

Electoral Systems

By comparing centralized (e.g., FPTP) and decentralized systems like proportional representation, the analysis highlights institutional variation as a key determinant of censorship levels.