Promoting the two-way flow of factors and achieving urban-rural integration is crucial in promoting the rational allocation of resources and narrowing the urban-rural gap. This paper constructs a conceptual model and empirically examines the impact of urban-rural integration on resource misallocation. It starts from the classic theories of development economics to sort out the logical mechanism of urban-rural integration in alleviating market distortions and improving resource allocation. Based on the panel data of prefecture-level cities in 11 provinces including National Pilot Zones for Integrated Urban-Rural Development, this paper calculates the resource misallocation index, applies Causal Forest and Synthetic DID to evaluate the impact of integrated urban-rural development on resource misallocation, and uses quantile regression to explore the heterogeneous impact.

The results show that this policy mainly alleviates labor misallocation but not capital misallocation. The alleviation in labor misallocation exhibits significant heterogeneity, primarily manifested in the most severe areas and the least severe areas. Channels of the effect on labor misallocation are mainly reflected in the emergence of new economic sectors and the improvement of Internet infrastructure. However, channels of the effect on capital misallocation are mainly due to irreversible investment, so the use of idle land to obtain funds will form a shock. The test also rules out the alternative explanation of cross-regional population mobility.

The contributions of this paper are as follows: First, from the perspective of urban-rural integration, it establishes a connection between resource allocation and the urban-rural relationship, investigating the empirical evidence of the integrated urban-rural development mechanism in alleviating resource misallocation. Second, based on the urban-rural characteristics implied by the supply elastic of labor factors, it constructs a mathematical model including the correlation between integrated urban-rural development, market distortion effects, and resource misallocation, extending the traditional resource misallocation research to the field of urban-rural relations. Third, it integrates machine learning technology into the identification strategy to effectively mitigate interference from endogeneity bias in empirical results.