This paper re-investigates the stationarity properties of per capita carbon dioxide (CO2) emissions and real Gross Domestic Product (GDP) per capita for 109 countries within seven regional panel sets covering 1971-2003. We apply the recent unit-root test of the panel seemingly unrelated regressions augmented Dickey-Fuller (SURADF) test developed by Breuer et al. [2001. Misleading inferences from panel unit-root tests with an illustration from purchasing power parity. Review of International Economics 9, 482-493; 2002. Series-specific unit-root tests with panel data, Oxford Bulletin of Economics and Statistics 64, 527-546]. The panel SURADF test accounts for the presence of cross-country correlations in the data, and the parameters in the panel specification vary across countries. More importantly, this test allows us to identify how many and which members of the panel contain a unit root. Overall, our empirical results illustrate that real GDP and CO2 emissions in these countries are a mixture of l(0) and l(1) processes, and that the traditional panel unit-root tests could lead to misleading inferences as well as the conduct of cointegration analysis being perhaps inappropriate. The results of our analysis carry critical implications for the modeling of CO2 emissions and GDP because of the different orders of integration for the two variables. (C) 2008 Elsevier Ltd. All rights reserved.