Abstract: Based on the panel data of 26 provinces in China from 1997 to 2017,this paper conducts an empirical analysis of the relationship between government health expenditure and health care consumption of urban and rural residents by constructing a panel threshold regression model.The results show that taking residents’income as a threshold variable,both the national and urban samples have a threshold and can pass the robustness test,indicating that government health expenditure will squeeze into residents’health care consumption,and the crowding effect will decreases with the increase of residents’income;while the rural samples do not have the threshold effect of residents’income.When further dividing the regions according to the urban residents’income threshold,the result shows that the income of urban residents in the central and western regions of China is at a lower level and the government health expenditure has a strong crowding-in effect,while the income of urban residents in eastern coastal areas is at a higher level and the government health expenditure has a weak crowding-in effect.

Key words: government health expenditure, residents' health care consumption, the threshold effect, residents income, crowding-in effects