More Hats, More Innovation? Nonlinear Effect of Accumulating Pilot Designations on Urban Innovation in China

• In Event: Geography, Networks, and Regional Innovation

Abstract

China's experimentalist governance has long relied on pilot policies to test and scale institutional reforms. Yet cities are increasingly becoming sites of policy accumulation, receiving multiple concurrent science and technology innovation (STI) piliot designations from the central government. Does stacking more “hats” on a city consistently enhance its innovation performance, or does a tipping point exist beyond which policy crowding undermines performance? This study investigates the nonlinear relationship between the number of STI pilot designations a city holds and its innovation output, seeking to identify the optimal policy density. We integrate Signaling Effect Theory with Institutional Innovation Ecosystem Theory to frame pilot designations as “institutional nutrients” fed into a city's innovation ecosystem. A moderate policy supply enriches resource diversity and sends credible pro-innovation to market actors, attracting talents capital, and entrepreneurial activity. However, excessive designations may overwhelm the ecosystem's absorptive capacity, leading to attention fragmentation among local officials, administrative overload from competing mandates, and ultimately diminishing returns.Empirically, we constructed a balanced panel dataset of 293 Chinese cities spanning 2007-2023. The key independent variable is the cumulative count of national STI pilot designations (e.g., Innovative City, Smart City, Intellectual Property Demonstration City) assigned to each city. Urban innovation performance, the dependent variable is measured by the ratio of patent granted to GDP. A quadratic regression specification is employed to capture the nonlinearity.The empirical results reveal a statistically significant inverted U-shaped relationship. Innovation output rises with each additional designation up to a turning point of approximately six concurrent pilots, after which the marginal effect turns negative. In other words, the first several “hats” substantially boost a city’s innovation, but the seventh and beyond begin to erode those gains, suggesting a policy overload effect. Heterogeneity analysis further highlight distinct spatial and hierarchical variations. The inverted U-shape is highly robust among provincial capitals, with 24 out of 27 capitals (88.89%) individually exhibiting this curvilinear pattern. Regionally, cities in China's eastern and central capitals have largely crossed the estimated threshold, meaning additional policy designations now yield diminishing or even negative marginal returns. Conversely, many western capitals remain on the upward slope of the curve, possessing ample absorptive capacity to translate further policy accumulation into innovation gains before hitting saturation.These findings challenge the implicit “more-is-better” logic prevalent in China’s top down policy diffusion, and offer robust estimate of a policy saturation threshold in the STI domain. Theoretically, the study extends ecosystem and signaling frameworks to explain non-monotonic policy effects. Practically, it offers actionable guidance for central authorities seeking to coordinate cross-sectoral pilot assignments to avoid redundant layering. Local governments would benefit from prioritize the depth of policy implementation over the breadth of designation, thereby enhancing the overall efficiency of China’s national system.

Author

• Xinle Cai, Huazhong University of Science and Technology
  Presenting Author