Globalization has led countries to a strong interdependence among them, which is reflected in trade and capital flows. Simultaneously, in recent decades, the world is rapidly urbanizing. This dynamic has generated a process of economic growth with serious consequences for the environment, particularly in air quality. In this context, the objective of this research is to examine the causal link among carbon dioxide emissions per capita as a measure of air pollution, real per capita output, FDI, trade openness, and urbanization in 100 countries during 1980-2017. First, we used the cointegration test of Pedroni (JAMA 61:653-670, 1999) and Westerlund (JAMA 69:709-748, 2007) to find the equilibrium long and short term, respectively, and the Dumitrescu and Hurlin (JAMA 29:1450-1460, 2012) test to verify the direction of causality among the series. Second, we estimate the strength of the cointegration vector for individual countries through a dynamic ordinary least squares model (DOLS), and for country groups using a dynamic panel model with ordinary least squares (PDOLS). The results found indicate the existence of short- and long-term equilibrium among the variables globally and by groups of countries. The strength of the cointegration vector is strong in high and middle-high-income countries. At a global level, the results of the causality test suggest the existence of a unidirectional causal relationship that goes from output, urbanization, and FDI to air pollution, and a bidirectional relationship among trade and air pollution. These results are sensitive to the inclusion of the level of development of the countries. Our results suggest that the mechanisms to increase output, along with commercial and FDI flows, and urbanization are factors that play a relevant role in the determination of air pollution. Consequently, public policies should take these aspects into account in efforts to mitigate air pollution.
Keywords: Causality; Cointegration; Environmental pollution; Output; Panel data.