Density, Cities and Air Pollution: A Global View

In this paper, we take a global view at air pollution looking at countries and cities worldwide. In doing so, we revisit the relationship between population density and air pollution, using i) a large panel of countries with data from 1960 to 2010, and ii) a unique and large sample of more than 1200 (big) cities around the world, combining pollution data with satellite data on built-up areas, population and light intensity at night at the grid-cell level for the last two decades. At the country level, we find that higher density in urban areas is associated with lower CO2 and PM2.5 emissions per capita. This result is supported at the city level; denser cities show lower emissions per capita. Our finding is robust to several controls and different estimation techniques and identification strategies. In our city level analysis, we also investigate the role of various characteristics of cities, in particular their average income, size and spatial structure (indicating within-city differences in density). We find evidence of an Environmental Kuznets Curve between economic development and pollution and that a polycentric city structure leads to lower pollution in the largest urban areas, while monocentricity seems beneficial for smaller cities.