Stochastic convergence of per capita carbon dioxide emissions and multiple structural breaks in OECD countries

This research applies an innovative panel data stationarity testing procedure developed by Carrion-i-Silvestre et al. [Carrion-i-Silvestre, J.L, Barrio-Castro, T.D. and Lopez-Bazo, E., 2005. Breaking the panels: An application to the GDP per capita, Econometrics journal 8, 159-175.], which has the advantage of recognizing multiple structural breaks and the presence of cross-section dependence in order to re-investigate the hypothesis that per capita carbon dioxide (CO2) emissions stochastically converge for 21 OECD countries from 1950 to 2002. Remarkably, the evidence clearly indicates that the panel dataset of relative per capita CO2 emissions is stationary after the structural breaks and cross-sectional dependence are introduced into the model. These findings offer strong policy implications for governments, regardless of whether they are in "convergent group" or "divergent group" countries. We also find that the structural breaks in the 1960s and over the 1970-1982 period are associated with time periods of fossil fuel becoming the main source of productivity, higher oil prices, and the development of nuclear power. (C) 2009 Elsevier B.V. All rights reserved.