Based on the panel data of 285 prefecture-level cities and above in China from 2005 to 2019, this paper takes the "Broadband China" pilot policy as a quasi-natural experiment and evaluates the impact of Internet development on urban eco-efficiency (symbolized by the "Broadband China" policy) by constructing multi-period difference-in-difference (DID) and spatial DID models. Results show that: the "Broadband China" pilot policy significantly improves the urban eco-efficiency: the eco-efficiency in pilot cities is about 16.8% higher than that in other cities. The results remain consistent after testing for robustness, including using estimation methods, excluding the sample of key cities, changing core explanatory variables, and introducing instrumental variables. Next, the influence of the "Broadband China" pilot policy on eco-efficiency is characterized by significant regional heterogeneity: Internet development significantly improves the eco-efficiency in the central, eastern and northeastern regions that are economically more developed and not resource-dependent. In contrast, this effect is not obvious in the western region that is economically less developed and resource-dependent. Moreover, the influencing mechanism of Internet development on eco-efficiency suggests that the "Broadband China" strategy boosts urban eco-efficiency by increasing the Internet penetration rate, improving technological innovation capacity, and upgrading the industrial structure. In addition, results from the spatial DID models indicate that the "Broadband China" pilot policy improves the eco-efficiency in local cities and significantly enhances that in neighboring cities. Based on this, this paper puts forward some suggestions regarding promoting new network infrastructure construction and differentiating development policies to fit local conditions.
Keywords: Internet; difference-in-difference method; eco-efficiency; “Broadband China” pilot policy.