The digital economy's rapid expansion in China, from 11 trillion yuan in 2016 to 45.5 trillion yuan in 2023, signifies its crucial role in driving high-quality economic growth. The government's "Digital China" strategy aims to deeply integrate the digital and real economies, fostering innovation-driven development. This shift from investment-driven to innovation-driven growth necessitates a thorough understanding of the digital economy's influence on innovation. The digital economy, characterized by big data and the internet, differs significantly from previous industrial revolutions, emphasizing the importance of data in regional economic development. This paper, therefore, focuses on analyzing the impact of the "Digital China" strategy on urban innovation and the mechanisms involved, providing valuable insights for sustainable urban development and policy implementation.

Existing research on the digital economy's impact on innovation faces challenges in measuring the digital economy itself, with varying sub-indicators leading to inconsistent composite indicators. Studies have explored the effects of digital access, equipment, platform construction, and application, revealing varied relationships with innovation. Some studies focus on informatization and internet development, while others use provincial or industry-level data. However, there's a scarcity of research holistically evaluating the digital economy's impact on urban innovation using policy evaluation. Existing studies mainly focus on the localized "Broadband China" policy, neglecting the broader scope of the "Digital China" strategy. This study addresses this gap by employing a multi-temporal DID approach using the localized implementation of digital economy strategies as a quasi-natural experiment to examine the impact, mechanisms, and heterogeneity of the digital economy's effect on urban technological innovation.

To address the endogeneity issue, this paper uses a multi-temporal double-difference (DID) method with the urban "Digital China" strategy as a quasi-natural experiment. This approach compares changes in urban innovation levels in treatment (cities with implemented strategies) and control groups before and after policy implementation. The core explanatory variable is the interaction term of whether a city's province implemented a digital economy policy and whether the year is after the policy implementation. The study uses panel data from 286 prefecture-level cities in China from 2008 to 2018. The dependent variable is the number of patent applications per 10,000 people. Control variables include GDP, population, industrial structure, fiscal capacity, investment, and science and technology expenditure. A parallel trend test is conducted to ensure the validity of the DID approach. The study also examines the mediating role of industrial structure upgrading using a mediation analysis. Further robustness tests are conducted by considering sample selectivity, excluding potentially confounding policies, substituting explanatory and control variables, and performing placebo tests. Finally, heterogeneity analysis is conducted based on innovation types (substantive vs. non-substantive) and regional location (East-Central vs. Western regions).

The benchmark regression analysis shows a significant positive impact of the "Digital China" strategy on urban innovation. The coefficient suggests that the strategy increases patent applications per 10,000 people by 2.32. The parallel trend test confirms that the observed effect is not due to pre-existing trends. Robustness tests, including sample adjustments, policy exclusions, variable substitutions, and placebo tests, consistently support the positive and significant effect of the "Digital China" strategy on urban innovation. Mechanism analysis reveals that industrial structure upgrading plays a significant mediating role in this relationship. Heterogeneity analysis indicates a stronger positive effect of the strategy on substantive innovation (invention patents) compared to non-substantive innovation. Furthermore, the strategy's impact on urban innovation is more pronounced in the East-Central regions than in the Western regions.

The findings strongly support the hypothesis that the digital economy, particularly through the "Digital China" strategy, significantly enhances urban innovation in China. The mediating effect of industrial structure upgrading suggests that policies promoting digital industrialization and industrial digitization are crucial for boosting innovation. The greater impact on substantive innovation highlights the importance of policies that foster high-quality innovation. The regional heterogeneity indicates that a differentiated approach to policy implementation is necessary to address regional disparities. These findings contribute to a better understanding of the digital economy's role in innovation and offer valuable insights for policymakers.

This study provides robust empirical evidence demonstrating the positive impact of the "Digital China" strategy on urban innovation in China, particularly through industrial structure upgrading and promotion of high-quality innovation. Policy recommendations include accelerating the strategy's implementation, focusing on foundational technologies, enhancing industrial policies, and addressing regional disparities through differentiated strategies. Future research could explore micro-level data analysis, develop theoretical models of digital economy and innovation, and explore more nuanced measures of technological innovation.

The study relies on patent applications as a proxy for innovation output, which might not fully capture all aspects of innovation. Data availability limitations restrict the analysis to prefecture-level city data, preventing a more granular analysis at the firm or individual level. The focus is on China, limiting the generalizability of the findings to other contexts with different institutional settings and economic structures.