Haze pollution not only has a huge impact on economic development but also seriously damages the health of residents, which has attracted the attention of many countries and scholars. The geographical detector model and the panel quantile regression model are used in combination to analyze the socio-economic driving factors of haze pollution from 2010 to 2015 for 74 significantly representative countries. The main results are as follows: (1) Industrial structure is the main factor affecting the haze concentration, followed by economic growth and research and development (R&D) intensity. (2) Government influence and industrial structure will significantly aggravate haze pollution, whereas the energy intensity and economic growth have an inhibitory effect on haze concentration. Countries with severe haze pollution should focus on upgrading their industrial structure and avoiding energy rebound. (3) Urbanization, foreign investment, and R&D intensity have different effects on the haze concentration among countries with different pollution levels. Specifically, the relationship between economic growth and pollution is inverted N-shaped in countries with medium haze concentration, whereas in other countries, it is positive N-shaped. Countries should actively leverage the agglomeration effect of high-density urban populations and focus on the introduction of high-quality foreign capital.
Keywords: Geographical detector model; Government influence; Haze pollution; Panel quantile regression.