From co-location to development: Spatial co-location of key digital technologies and urban economic performance in the Yangtze River Delta

As digital technologies become embedded in urban production and governance, high-quality urban development depends not only on growth in technology volume but also on whether key digital technologies form stable spatial co-location and complementary configurations within cities. This study examines 41 cities in the Yangtze River Delta, utilizing CNIPA-issued patents with enterprise applicants from 2000 to 2024. Patents are classified into seven key digital technology types, and multi-order co-location patterns are identified from point-level neighborhood relations among geocoded patent applicant locations within each city-year unit. These point-based co-location results are then aggregated into city-year configuration indicators for comparison with economic performance. The analysis evaluates both contemporaneous and lagged links with city-level economic performance, and applies year-grouped cross-validation with an interpretable predictive model to identify influential configurations and their marginal contributions using XGBoost-SHAP. The results show that city economic performance aligns less with the simple expansion of patent counts or technology categories than with stable and increasingly complex cross-technology co-location configurations. Configuration dynamics combine persistence with adjustment: a small set of combinations remains repeatedly influential in predictive models and is associated with later expansion and recombination. The model-based associations exhibit both an intensity threshold and a time lag: higher and more stable co-location intensity is more often associated with positive predicted economic differences. These findings suggest that digital policy and industrial spatial planning may benefit from paying greater attention to cross-technology configuration capacity. At the same time, the evaluation should consider the time process through which spatial co-location is absorbed and converted into economic outcomes.