The world distribution of carbon dioxide emissions

Abstract

This article examines the spatial distribution of per capita carbon dioxide emissions in 140 countries during the period 1960 to 1999. To overcome the methodological limitations of conventional convergence analysis, we have applied a nonparametric approach, which allows to study the dynamics of the entire cross section distribution. The results show that the probability mass concentrated around the average has increased over time, which helps to explain the observed reduction in cross-country disparities. In any event, the intra-distribution mobility level is relatively low.

Notes

¹ The data used are from the World Development Indicators (World Bank, 2003) and involve CO₂ emissions from the burning of fossil fuels and the manufacture of cement. The entire list of countries considered is given in the Appendix.

² When interpreting Fig. 1, it should be noted that we normalized each country's level of per capita CO₂ emissions according to the average of the distribution for the year in question. The average was assigned a value of 100. This enabled us to work with relative values, thus removing the influence of global changes in emission levels over time.

³ For a formal definition of these instruments, see Durlauf and Quah (1999).

⁴ To test the robustness of these findings, we estimated the stochastic kernels for 10- and 15-year horizons. The results are very similar to those in Fig. 2 and are available upon request.