Objective We examine empirical evidence of region‐based environmental injustice in China, and factors, especially environmental public service, that contribute to the distribution of environmental hazards among various Chinese regions. Methods We use variance analysis to assess the disparate distribution of environmental hazards, and then employ the expanded Stochastic Impacts by Regression on Population, Affluence, and Technology (STIRPAT) model to identify the determinants, using a panel data set of 26 provincial capital cities and four municipalities directly under Chinese central government over the period 2003–2010. Results and Conclusions Our findings confirm the presence of region‐based environmental injustice in China and demonstrate its annual improvement. We also find that environmental public service, economic development level, population size, and environmental protection technology are major impact factors. These findings not only support the environmental Kuznets curve assumption, but also help to advance the existing literature on environmental justice and the human‐environment relationship.