Is China’s Urbanization Inclusive?—Comparative Research Based on Machine Learning Algorithms

Abstract

Improving the inclusiveness of urban development is crucial to improving the wages of low- and middle-income workers. In this study, we used machine learning to cluster urban labor into low, middle, and high socioeconomic groups in order to analyze the effects of economic agglomeration and compare them with the results, which were classified according to income. The results showed that economic agglomeration has improved the wages of the low and middle socioeconomic groups; the estimated wage spillover effect was 3.9%. By contrast, the estimated result based on the groups classified by a single index of income was 20.3%, which represents an overestimation of the wage spillover effect of economic agglomeration. This method is often used to explain the inclusiveness of China’s urbanization, leading to overestimation. Further mechanism analysis found that the characteristics of the industrial structure affect the change in wage elasticity caused by economic agglomeration, which has a moderating effect on the wages of workers. The differing wage elasticity associated with economic agglomeration is responsible for wage disparities in China.