From Network Deployment to AI Industry Location: How Mobile Broadband Shapes the County-Level Geography of AI Firms in China
Abstract Digital networks reduce the costs of communication and coordination across distance, but their spatial consequences remain ambiguous. This study examines whether mobile broadband allows artificial intelligence (AI) firms to escape local constraints or instead reconfigures county-level locational advantage by improving the complementary conditions needed for financing, staffing, and innovation. We assemble a balanced panel of 1,866 county-level administrative units in China from 2015 to 2022. Geocoded LTE base-station records from OpenCelliD are matched to county boundaries to measure deployment intensity, and Tianyancha business-registration data are used to identify AI firms and construct their density per square kilometer. Two-way fixed-effects estimates show that more intensive observed LTE deployment is associated with greater AI-firm density. Instrumental-variable estimates and multiple robustness checks provide additional support. Mechanism-consistent evidence indicates that deployment is associated with broader digital financial inclusion and stronger bank competition, greater digital-labor supply and recruitment demand, and denser university-industry and interfirm innovation links. The association is stronger in non-resource-dependent areas, localities with denser road networks, and counties with greater economic resilience. The findings suggest that mobile broadband does not erase geography. It changes the basis of locational advantage by improving how complementary resources are identified, matched, and coordinated.
Authors
- Junjie Ruan
- Tian Cheng
Institutions
- Zhongnan University of Economics and Law (CN)
- Guilin University of Electronic Technology (CN)
Publication Details
- Journal
- Research Square
- Published
- 2026-11-17
- DOI
- https://doi.org/10.21203/rs.3.rs-10961050/v1
- Primary Topic
- Agglomeration and Urban Systems
- Type
- preprint