The effect of natural and anthropogenic factors on PM_(2.5): Empirical evidence from Chinese cities with different income levels

摘要

The aim of this paper is to estimate the effects of natural conditions and anthropogenic factors on PM2.5 concentrations, taking into consideration differences in the income levels, and thus the development stages, of the cities studied. To achieve this goal, a balanced dataset of 287 Chinese cities was divided into different income-based panels for the period 1998-2015. The empirical estimation results indicated that meteorological conditions exerted varied effects on PM2.5 concentrations across different income-based panels. The results show that the coefficients of temperature were positive and significant in all panels, with the exception of upper-middle-incomecities. Whilst wind speed and precipitation were found to be conducive to reducing PM2.5 concentrations, no such significant correlation was found in relation to relative humidity (except in high-income cities). In terms of the anthropogenic factors addressed in the study, we found an inverted U-shaped relationship between economic development and PM2.5 concentrations, confirming the Environmental Kuznets Curve hypothesis. In addition, the industrial structure and road density were observed to exert significant positive impacts on PM2.5 concentrations. The empirical analysis of the effects of FDI on PM2.5 concentrations indicate that FDI aggravated PM2.5 pollutions in the total cities and lower-middle-income cities panels, supporting the Pollution Haven Hypothesis. The empirical results for population density suggested that it does not significantly influence PM2.5 concentrations. Moreover, we found that built-up area exerts mixed effects on PM2.5 concentrations. These results cast a new light on the issue of PM2.5 pollution for government policy makers tasked with formulating measures to mitigate the concentration of such pollutants, encouraging that consideration be given to the differences between cities with different income levels. (c) 2018 Elsevier B.V. All rights reserved.