Public data openness and urban innovation: evidence from a quasi-natural experiment in Chinese cities

Does releasing public data make cities more innovative—or does it only help cities that are equipped to use it? Open data is widely promoted as a driver of innovation, yet evidence on its causal effects, and conditions is scarce. We address this gap using the staggered rollout of municipal public data open platforms across 280 Chinese prefecture-level cities (2007–2021) as a quasi-natural experiment. Employing a two-way fixed effects difference-in-differences design, we find that public data openness significantly increases urban innovation capacity—robust across the Callaway–Sant’Anna estimator, placebo tests, PSM-DID, and a continuous policy-intensity measure. Two complementary mechanisms—enhanced intercity knowledge flows and digital talent agglomeration—receive support from first-stage estimation and bootstrap mediation tests. Strikingly, the policy benefits cities with lower education levels as much as wealthier ones, yet leaves western cities unaffected. We propose a unified framework to reconcile this contradiction: low-education cities face a demand-side information gap that open data fills, while western cities face a supply-side capacity deficit that data release alone cannot overcome. This reframes how policymakers should think about open data—not as a uniform solution, but as a policy whose payoff is conditional on what a city is missing.