Monopolizing Transparency: Engineering Automated Air Quality Data in China

Government deployment of automated monitoring networks -- sensors that collect and disseminate public data -- is widely celebrated as a technological solution for transparency and accountability. Yet these technologies depend on physical infrastructure that, once installed, becomes largely fixed while political priorities often change. I argue that this strategic tension incentivizes infrastructure designs that filter unfavorable information from the outset, allowing both politicians and implementing agencies to hedge uncertainty. Analyzing the expansion of China's automatic ambient air quality monitoring network (2012-2022) through original geo-referenced datasets on monitoring locations, pollution sources, and satellite imagery, I document systematic distortions: monitors are underdeployed by 8.7% in polluted areas, 11% more likely to be placed in greener environments, and experience 2% more downtime during pollution spikes. These findings illustrate how infrastructure configuration serves as a mechanism that undermines information integrity before data collection even begins, creating an accountability gap hidden beneath a veneer of technical neutrality. --- Event Details: https://my.princeton.edu/rsvp?id=1960832