Informal Institutions and Authoritarian Information Systems : Theory and Evidence from China

Authoritarian regimes face pervasive information problems, not only between state and society but also inside of their own hierarchies. How do authoritarian leaders address internal information issues? We argue that authoritarian information systems contain both formal and informal channels and show substitution between them using evidence from China's economic reporting system. We construct a new measure for economic falsification by subtracting a “plausible growth rate” generated by the Random Forests technique from the official GDP growth rate for each of China's cities. We find that growth over-reporting is less severe in cities where higher-level political leaders enjoy strong informal networks that can serve as alternative information channels. This result is robust across different falsification measures and various alternative specifications. We also explore heterogeneity and spillovers associated with this distortion-reduction effect. Our findings underscore the importance of informal institutions and contribute to a more comprehensive understanding of authoritarian information systems