Identifying Environmental Kuznets Curves: The Case of SO 2 and CO 2 emissions ∗ Herman R.J. Vollebergh † Bertrand Melenberg ‡ Elbert Dijkgraaf § This version: April 1, 2008 Abstract In this paper we offer a solution to the non-robustness that plagues reduced form estimations of the inverted-U shaped relationship between emissions and income. Reduced form estimations are very sensitive to the specification of the time controls due to identification problems. This paper develops a new estimation framework that explicitly distinguishes between what can be iden- tified on the basis of the data and what is a consequence of subjective choices due to a lack of identification. We apply our methodology to two widely stud- ied cases, namely, SO 2 and CO 2 emissions. Interestingly, our estimates are insensitive to the required subjective choices, but also strongly differ from the literature so far. We find consistent positive income effects for both cases and time effect estimates with a clear U-shaped trend for SO 2 -emissions but only slightly so for CO 2 -emissions. Keywords: Identification; Panel Data; Reduced-Form (Semi-)Parametric Estimation; Emission-Income Relationships JEL Codes: C33; O50; Q40 ∗ We thank Arik Levinson, Daniel Millimet, and Aart de Zeeuw for constructive comments. Earlier versions of this paper also benefited from discussions with Otto Swank, Scott Taylor, and seminar participants at the Sustainable Resource Use and Economic Dynamics conference in Ascona 2004, Econometrics and Statistics seminar at Tilburg University, Econometric Society European Meetings, as well as Third World Congress of Environmental and Resource Economists in Kyoto 2006. † Corresponding author, Tinbergen Institute, Erasmus University Rotterdam, H 8-1, P.O. Box 1738, 3000 DR Rotterdam, The Netherlands, vollebergh@few.eur.nl ‡ CentER and Department of Econometrics, Tilburg University § SEOR-ECRi, Tinbergen Institute, Erasmus University Rotterdam 1